Changes between Version 38 and Version 39 of GENIBibliography


Ignore:
Timestamp:
05/05/16 13:15:34 (5 years ago)
Author:
Mark Berman
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • GENIBibliography

    v38 v39  
    3838<hr>
    3939<a id="full"><H1>GENI Bibliography</H1></a>
    40 Year:&nbsp;
    41 
    42 <a href="#full-2006">2006</a>&nbsp;
    43 
    44 <a href="#full-2007">2007</a>&nbsp;
    45 
    46 <a href="#full-2008">2008</a>&nbsp;
    47 
    48 <a href="#full-2009">2009</a>&nbsp;
    49 
    50 <a href="#full-2010">2010</a>&nbsp;
    51 
    52 <a href="#full-2011">2011</a>&nbsp;
    53 
    54 <a href="#full-2012">2012</a>&nbsp;
    55 
    56 <a href="#full-2013">2013</a>&nbsp;
    57 
    58 <a href="#full-2014">2014</a>&nbsp;
    59 
    60 <a href="#full-2015">2015</a>&nbsp;
    61 
    6240<br>
    6341<a href="#concise">Jump to concise bibliography</a>
    6442<ol>
    65 
    66 <br>
    67 <a id="full-2006"><H2>GENI Publications for 2006</H2></a>
     43<br>
     44
     45
     46<li>
     47<b>&#x4f;&#x0308;z&#x63;&#x0327;elik, &#x49;&#x0307;lker and Brooks, Richard R.</b>
     48, &quot;Deceiving entropy based DoS detection.&quot;
     49Computers &#x0026; Security,
     502015.
     51doi:10.1016/j.cose.2014.10.013.
     52<a href="http://dx.doi.org/10.1016/j.cose.2014.10.013">http://dx.doi.org/10.1016/j.cose.2014.10.013</a>
     53<br><br><b>Abstract: </b>Denial of Service (DoS) attacks disable network services for legitimate users. As a result of growing dependence on the Internet by both the general public and service providers, the availability of Internet services has become a concern. While DoS attacks cause inconvenience for users, and revenue loss for service providers; their effects on critical infrastructures like the smart grid and public utilities could be catastrophic. For example, an attack on a smart grid system can cause cascaded power failures and lead to a major blackout. Researchers have proposed approaches for detecting these attacks in the past decade. Anomaly based DoS detection is the most common. The detector uses network traffic statistics; such as the entropy of incoming packet header fields (e.g. source IP addresses or protocol type). It calculates the observed statistical feature and triggers an alarm if an extreme deviation occurs. Entropy features are common in recent DDoS detection publications. They are also one of the most effective features for detecting these attacks. However, intrusion detection systems (IDS) using entropy based detection approaches can be a victim of spoofing attacks. An attacker can sniff the network and calculate background traffic entropy before a (D)DoS attack starts. They can then spoof attack packets to keep the entropy value in the expected range during the attack. This paper explains the vulnerability of entropy based network monitoring systems. We present a proof of concept entropy spoofing attack and show that by exploiting this vulnerability, the attacker can avoid detection or degrade detection performance to an unacceptable level.
     54</li>
     55<br>
     56
     57
     58
     59<li>
     60<b>Aikat, Jay and Hasan, Shaddi and Jeffay, Kevin and Smith, F. Donelson</b>
     61, &quot;Discrete-Approximation of Measured Round Trip Time Distributions: A Model for Network Emulation.&quot;
     62First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     632012.
     64
     65
     66<br><br><b>Abstract: </b>Empirical evaluations to study network performance, whether in a laboratory setting or on GENI testbeds, rely heavily on measurement-based modeling of round trip times (RTTs) to emulate realistic end-to-end delays of local and metropolitan area networks. For generating realistic traffic, we studied several models to emulate RTTs. In this paper, we performed experiments on real testbeds using synthetic TCP traffic generated from measurement data from a large university campus. As a result of our study, we present the Discrete- Approximation model for RTT (DA-RTT) emulation. Using three different metrics for performance evaluation, which include queue length at routers, connection response times, and connection durations, we demonstrate that the simple DA-RTT model closely represents the per-connection RTTs in the original traffic. While these experiments were performed in our laboratory, and not using GENI infrastructure, we present this as a possible model for adoption on GENI testbeds to emulate Round Trip Time Distributions for GENI experiments.
     67</li>
     68<br>
     69
     70
     71
     72<li>
     73<b>Albrecht, J. and Huang, D. Y.</b>
     74, &quot;Managing distributed applications using Gush.&quot;
     75Proceedings of the ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities, Testbed Practices Session (TridentCom),
     762010.
     77doi:10.1007/978-3-642-17851-1&#x005F;31.
     78<a href="http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;31">http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;31</a>
     79<br><br><b>Abstract: </b>Deploying and controlling experiments running on a distributed set of resources is a challenging task. Software developers often spend a significant amount of time dealing with the complexities associated with resource configuration and management in these environments. Experiment control systems are designed to automate the process, and to ultimately help developers cope with the common problems that arise during the design, implementation, and evaluation of distributed systems. However, many of the existing control systems were designed with specific computing environments in mind, and thus do not provide support for heterogeneous resources in different testbeds. In this paper, we explore the functionality of Gush, an experiment control system, and discuss how it supports execution on three of the four GENI control frameworks.
     80</li>
     81<br>
     82
     83
     84
     85<li>
     86<b>Albrecht, Jeannie R.</b>
     87, &quot;Bringing big systems to small schools: distributed systems for undergraduates.&quot;
     88SIGCSE Bull., ACM, New York, NY, USA,
     892009.
     90doi:10.1145/1539024.1508903.
     91<a href="http://dx.doi.org/10.1145/1539024.1508903">http://dx.doi.org/10.1145/1539024.1508903</a>
     92<br><br><b>Abstract: </b>Distributed applications have become a core component of the Internet's infrastructure. However, many undergraduate curriculums, especially at small colleges, do not offer courses that focus on the design and implementation of distributed systems. The courses that are offered address the theoretical aspects of system design, but often fail to provide students with the opportunity to develop and evaluate distributed applications in real-world environments. As a result, undergraduate students are not as prepared as they should be for graduate study or careers in industry. This paper describes an undergraduate course in Distributed Systems that not only studies the key design principles of distributed systems, but also has a unique emphasis on giving students hands-on access to distributed systems through the use of shared computing testbeds, such as PlanetLab and GENI, and open-source technologies, such as Xen and Hadoop. Using these platforms, students can perform large-scale, distributed experimentation even at small colleges.
     93</li>
     94<br>
     95
     96
     97
     98<li>
     99<b>Albrecht, Jeannie and Tuttle, Christopher and Braud, Ryan and Dao, Darren and Topilski, Nikolay and Snoeren, Alex C. and Vahdat, Amin</b>
     100, &quot;Distributed application configuration, management, and visualization with plush.&quot;
     101ACM Trans. Internet Technol., ACM, New York, NY, USA,
     1022011.
     103doi:10.1145/2049656.2049658.
     104<a href="http://dx.doi.org/10.1145/2049656.2049658">http://dx.doi.org/10.1145/2049656.2049658</a>
     105<br><br><b>Abstract: </b>Support for distributed application management in large-scale networked environments remains in its early stages. Although a number of solutions exist for subtasks of application deployment, monitoring, and maintenance in distributed environments, few tools provide a unified framework for application management. Many of the existing tools address the management needs of a single type of application or service that runs in a specific environment, and these tools are not adaptable enough to be used for other applications or platforms. To this end, we present the design and implementation of Plush, a fully configurable application management infrastructure designed to meet the general requirements of several different classes of distributed applications. Plush allows developers to specifically define the flow of control needed by their computations using application building blocks. Through an extensible resource management interface, Plush supports execution in a variety of environments, including both live deployment platforms and emulated clusters. Plush also uses relaxed synchronization primitives for improving fault tolerance and liveness in failure-prone environments. To gain an understanding of how Plush manages different classes of distributed applications, we take a closer look at specific applications and evaluate how Plush provides support for each.
     106</li>
     107<br>
     108
     109
     110
     111<li>
     112<b>Angu, Pragatheeswaran and Ramamurthy, Byrav</b>
     113, &quot;Experiences with dynamic circuit creation in a regional network testbed.&quot;
     1142011 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Shanghai, China, IEEE,
     1152011.
     116doi:10.1109/infcomw.2011.5928801.
     117<a href="http://dx.doi.org/10.1109/infcomw.2011.5928801">http://dx.doi.org/10.1109/infcomw.2011.5928801</a>
     118<br><br><b>Abstract: </b>In this paper we share our experiences of enabling dynamic circuit creation in the GpENI network. GpENI is a network research testbed in the mid-west USA involving several educational institutions. University of Nebraska-Lincoln is involved in provisioning dynamic circuits across the GpENI network among its participating universities. We discuss several options investigated for deploying dynamic circuits over the GpENI network as well as our demonstration experiments at the GENI engineering conferences. UNL has also collaborated with ProtoGENI project of University of Utah and Mid-Atlantic Crossroads (MAX) facility of Washington DC to create inter-domain dynamic circuits.
     119</li>
     120<br>
     121
     122
     123
     124<li>
     125<b>Antonenko, V. and Smeliansky, R. and Baldin, I. and Izhvanov, Y. and Gugel, Y.</b>
     126, &quot;Towards SDI-bases Infrastructure for supporting science in Russia.&quot;
     127Science and Technology Conference (Modern Networking Technologies) (MoNeTeC), 2014 First International, IEEE,
     1282014.
     129doi:10.1109/monetec.2014.6995576.
     130<a href="http://dx.doi.org/10.1109/monetec.2014.6995576">http://dx.doi.org/10.1109/monetec.2014.6995576</a>
     131<br><br><b>Abstract: </b>Modern science presents a number of challenges to the cyber-infrastructure supporting it: heterogeneity of the required computational resources, problems associated with storing, preserving and moving large quantities of information, a collaborative nature of scientific activities requiring shared access to resources, continuously growing requirements for computational power and network bandwidth, and, last, but not least, ease of use. In this position paper we explore a new approach to creating and growing such infrastructure based on the principles of federation, enabled by deep programmability of individual infrastructure elements: Software-Defined Infrastructure (SDI). We describe the evolution of the science infrastructure, open research problems and the concrete steps we are taking towards its realization by building a unique, widely distributed science facility in Russia based on SDI and GENI technologies.
     132</li>
     133<br>
     134
     135
     136
     137<li>
     138<b>Araji, B. and Gurkan, D.</b>
     139, &quot;Embedding Switch Number, Port Number, and MAC Address (ESPM) within the IPv6 Address.&quot;
     140Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     1412014.
     142doi:10.1109/gree.2014.20.
     143<a href="http://dx.doi.org/10.1109/gree.2014.20">http://dx.doi.org/10.1109/gree.2014.20</a>
     144<br><br><b>Abstract: </b>IPv4 protocol, the famous 32-bit address, has been used in networks for many decades [1] and would not have sustained its usability without NAT. IPv6 protocol with its 128-bit address, provides slight routing information [2]. In this paper, we present ESPM, Embedding Switch ID, Port number and MAC Address within IPv6 protocol and SDN technology, imposing a device connectivity hierarchy upon the address space. We amend the IPv6 global addressing scheme for hosts to include their MAC address as well as the switch ID and Switch port number that they are connected to. This scheme encodes information that would ordinarily require a lookup or query packets(ARP) and decrease CAM table entries on the switch by forwarding the packets using the ESPM algorithm. After processing ESPM algorithm to check for OF controller ID, OF switch ID, and the port ID, the amount of total packets transferred on the network to fulfill an ICMP request-reply process decreased by 28.1&#x0025; in 1-switch-2 hosts. In order to demonstrate the feasibility of such an addressing scheme, we use POF controller and POF switch [3] to emulate ESPM implementation and then measure the impact on the number of network management packets transferred between hosts during connectivity tests.
     145</li>
     146<br>
     147
     148
     149
     150<li>
     151<b>Aug&#x65;&#x0301;, Jordan and Parmentelat, Thierry and Turro, Nicolas and Avakian, Sandrine and Baron, Lo&#x69;&#x0308;c and Larabi, Mohamed A. and Rahman, Mohammed Y. and Friedman, Timur and Fdida, Serge</b>
     152, &quot;Tools to foster a global federation of testbeds.&quot;
     153Computer Networks,
     1542014.
     155doi:10.1016/j.bjp.2013.12.038.
     156<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.038">http://dx.doi.org/10.1016/j.bjp.2013.12.038</a>
     157<br><br><b>Abstract: </b>A global federation of experimental facilities in computer networking is being built on the basis of a thin waist, the Slice-based Federation Architecture (SFA), for managing testbed resources in a secure and efficient way. Its success will depend on the existence of tools that allow testbeds to expose their local resources and users to browse and select the resources most appropriate for their experiments. This paper presents two such tools. First, SFAWrap, which makes it relatively easy for a testbed owner to provide an SFA interface for their testbed. Second, MySlice, a tool that allows experimenters to browse and reserve testbed resources via SFA, and that is extensible through a system of plug-ins. Together, these tools should lower the barriers to entry for testbed owners who wish to join the global federation.
     158</li>
     159<br>
     160
     161
     162
     163<li>
     164<b>Babaoglu, A. C. and Dutta, R.</b>
     165, &quot;A GENI Meso-Scale Experiment of a Verification Service.&quot;
     166Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     1672014.
     168doi:10.1109/gree.2014.13.
     169<a href="http://dx.doi.org/10.1109/gree.2014.13">http://dx.doi.org/10.1109/gree.2014.13</a>
     170<br><br><b>Abstract: </b>In this work, we demonstrate the real world results of a verification service that verifies the performance of a set of network providers by measuring the user flows, using GENI experimental facility. We first give an overview of the architectural components and their interactions to enable such a verification capability. We then give the experiment setup details followed by the numerical results for various network measurement metrics and the evaluation of these results.
     171</li>
     172<br>
     173
     174
     175
     176<li>
     177<b>Babaoglu, Ahmet C.</b>
     178, &quot;Verification Services for the Choice-Based Internet of the Future (Doctoral dissertation).&quot;
     179
     1802014.
     181
     182<a href="http://www.lib.ncsu.edu/resolver/1840.16/9336">http://www.lib.ncsu.edu/resolver/1840.16/9336</a>
     183<br><br><b>Abstract: </b>The Internet has grown from its inception as a special-purpose internetwork into a general multi-purpose world-wide facility enabling education, commerce, governance, and societal communication, all in the space of a few decades. Over this time, and accelerating in the last decade or so, increasing demands and a growing variety of use cases are posing new challenges on the architecture prompting re-thinking and re-architecting of the network. One thread of research in such architectural considerations involves the issue of choice. The lack of alternative network services brings little economic incentive for the network service providers to make investments to deploy new technologies and improve the quality of their network services. In addition, most user flows goes through several providers, thus there is no effective mechanism in the current Internet to provide feedback to users about which provider is the cause of the performance problems they experience. One solution to these problems is to create a more competitive open market where providers can advertise their network services, and users can choose their desired set of network services to satisfy their needs. In this solution, the users have the option to choose another service if they are not satisfied. However, even in this solution, the root cause of the performance problems still can not be found and it brings us to the lack of a robust feedback capability. In this work, we investigate a solution to this fundamental missing piece of the In- ternet, the measurement and verification capability of the network services offered in the Internet, that indirectly pushes more responsibility to the network providers to fulfill their requirements for high quality services. Our work, while rooted in standard expectations of economic theory, is not in economics itself. Rather, it is in defining, designing, and realizing architectural entities and interactions in technical terms that can realize verification services essential to enabling such economic interactions. Our work is threefold; after giving a literature overview of the research on future Internet and Internet measurement, we first propose an architecture that defines the roles, interactions and design choices to enable a Choice-Based Verification Service. We then describe the results and analysis of a series of tests, which start with our work on measurement frameworks in wired and wireless environments and continue with the simulation, the mechanism introduced and the actual prototype of this work deployed into a real system, the GENI meso-scale testbed. Finally, we investigate and validate whether such informed choices with verification service actually lead to better overall results. We use energy-efficiency as a practical and useful domain for a case study and show the simulation results, which greatly increase the appeal of this work as applicable real-world network services.
     184</li>
     185<br>
     186
     187
     188
     189<li>
     190<b>Baldine, I.</b>
     191, &quot;Unique optical networking facilities and cross-layer networking.&quot;
     192Summer Topical Meeting, 2009. LEOSST '09. IEEE/LEOS,
     1932009.
     194doi:10.1109/LEOSST.2009.5226210.
     195<a href="http://dx.doi.org/10.1109/LEOSST.2009.5226210">http://dx.doi.org/10.1109/LEOSST.2009.5226210</a>
     196
     197</li>
     198<br>
     199
     200
     201
     202<li>
     203<b>Baldine, Ilia and Xin, Yufeng and Evans, Daniel and Heerman, Chris and Chase, Jeff and Marupadi, Varun and Yumerefendi, Aydan</b>
     204, &quot;The missing link: Putting the network in networked cloud computing.&quot;
     205in ICVCI09: International Conference on the Virtual Computing Initiative,
     2062009.
     207
     208
     209
     210</li>
     211<br>
     212
     213
     214
     215<li>
     216<b>Baldine, Ilia and Xin, Yufeng and Mandal, Anirban and Renci, Chris H. and Chase, Unc-Ch J. and Marupadi, Varun and Yumerefendi, Aydan and Irwin, David</b>
     217, &quot;Networked cloud orchestration: A GENI perspective.&quot;
     2182010 IEEE Globecom Workshops, Miami, FL, USA, IEEE,
     2192010.
     220doi:10.1109/GLOCOMW.2010.5700385.
     221<a href="http://dx.doi.org/10.1109/GLOCOMW.2010.5700385">http://dx.doi.org/10.1109/GLOCOMW.2010.5700385</a>
     222<br><br><b>Abstract: </b>This paper describes the experience of developing a system for creation of distributed linked configurations of heterogeneous resources (slices) in GENI. Our work leverages a number of unique architectural solutions (distributed architecture, declarative resource specifications, unique approach to slice instantiation) which is applicable to a wider set of problems related to autonomic co-scheduling and provisioning of heterogeneous networked resources. We discuss the architecture, the resource description mechanisms and some of the algorithms used to enable our system. We conclude with an analysis of a real experiment at allocating resources from multiple providers across a very wide geographic area (spanning Massachusetts, Illinois and North Carolina) to create a single private Layer 2 network connecting virtual machines on the campus of Duke University to a sensor testbed at University of Massachusetts, Amherst.
     223</li>
     224<br>
     225
     226
     227
     228<li>
     229<b>Baldine, Ilia and Xin, Yufeng and Mandal, Anirban and Ruth, Paul and Yumerefendi, Aydan and Chase, Jeff</b>
     230, &quot;ExoGENI: A Multi-Domain Infrastructure-as-a-Service Testbed.&quot;
     2318th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TRIDENTCOM 2012),
     2322012.
     233
     234
     235<br><br><b>Abstract: </b>NSF's GENI program seeks to enable experiments that run within virtual network topologies built-to-order from testbed infrastructure offered by multiple providers (domains). GENI is often viewed as a network testbed integration effort, but behind it is an ambitious vision for multi-domain infrastructure-as-a-service (IaaS). This paper presents ExoGENI, a new GENI testbed that links GENI to two advances in virtual infrastructure services outside of GENI: open cloud computing (OpenStack) and dynamic circuit fabrics. ExoGENI orchestrates a federation of independent cloud sites and circuit providers through their native IaaS interfaces, and links them to other GENI tools and resources. The ExoGENI deployment consists of cloud site ``racks'' on host campuses within the US, linked with national research networks and other circuit networks through programmable exchange points. The ExoGENI sites and control software are enabled for software-defined networking using OpenFlow. ExoGENI offers a powerful unified hosting platform for deeply networked, multi-domain, multi-site cloud applications. We intend that ExoGENI will seed a larger, evolving platform linking other third-party cloud sites, transport networks, and other infrastructure services, and that it will enable real-world deployment of innovative distributed services and new visions of a Future Internet.
     236</li>
     237<br>
     238
     239
     240
     241<li>
     242<b>Bashir, Sadia and Ahmed, Nadeem</b>
     243, &quot;VirtMonE: Efficient detection of elephant flows in virtualized data centers.&quot;
     244Telecommunication Networks and Applications Conference (ITNAC), 2015 International, IEEE,
     2452015.
     246doi:10.1109/atnac.2015.7366826.
     247<a href="http://dx.doi.org/10.1109/atnac.2015.7366826">http://dx.doi.org/10.1109/atnac.2015.7366826</a>
     248<br><br><b>Abstract: </b>A modern virtualized data center is highly multifarious environment shared among hundreds of co-located tenants hosting heterogeneous applications. The tenants' virtual machines generate a subset of elephants or mice flows (different in terms of rate, size, duration, and burstiness) based on the type of application they are running. Virtual traffic generated from the tenant's virtual machines traverses the underlay physical fabric in aggregate because of different encapsulation techniques (VXLAN, NVGRE, and STT for example) employed in data center networks thus obfuscating the virtual traffic characteristics. Existing approaches to monitor and/or identify elephant flows either have limited or no visibility into virtual traffic or are associated with high monitoring overhead making it hard to precisely detect and properly engineer elephant flows on the underlay fabric. In this paper, we present VirtMonE, a lightweight detection mechanism aimed at precisely detecting egress elephant flows at Open vSwitch while providing visibility into virtual traffic with least measurement and monitoring overhead at the edge. We conduct simulations on a small GENI testbed to evaluate the performance of the proposed solution for a software-defined multi-tenant virtual network. Our proposed solution is demonstrated to precisely detect the elephant flows from different tenants at the edge, provide visibility into virtual traffic and mitigate the network overhead associated with detection, thus improving the overall performance of the data centre.
     249</li>
     250<br>
     251
     252
     253
     254<li>
     255<b>Bastin, Nicholas and Bavier, Andy and Blaine, Jessica and Chen, Jim and Krishnan, Narayan and Mambretti, Joe and McGeer, Rick and Ricci, Rob and Watts, Nicki</b>
     256, &quot;The InstaGENI initiative: An architecture for distributed systems and advanced programmable networks.&quot;
     257Computer Networks,
     2582014.
     259doi:10.1016/j.bjp.2013.12.034.
     260<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.034">http://dx.doi.org/10.1016/j.bjp.2013.12.034</a>
     261<br><br><b>Abstract: </b>In this paper, we describe InstaGENI, a distributed cloud based on programmable networks designed for the GENI Mesoscale deployment and large-scale distributed research projects. The InstaGENI architecture closely integrates a lightweight cluster design with software-defined networking, Hardware-as-a-Service and Containers-as-a-Service, remote monitoring and management, and high-performance inter-site networking. The initial InstaGENI deployment will encompass 34 sites across the United States, interconnected through a specialized GENI backbone network deployed over national, regional and campus research and education networks, with international network extensions to sites across the world.
     262</li>
     263<br>
     264
     265
     266
     267<li>
     268<b>Bavier, Andy and Chen, Jim and Mambretti, Joe and McGeer, Rick and McGeer, Sean and Nelson, Jude and O'Connell, Patrick and Ricart, Glenn and Tredger, Stephen and Coady, Yvonne</b>
     269, &quot;The GENI experiment engine.&quot;
     270Teletraffic Congress (ITC), 2014 26th International, IEEE,
     2712014.
     272doi:10.1109/itc.2014.6932974.
     273<a href="http://dx.doi.org/10.1109/itc.2014.6932974">http://dx.doi.org/10.1109/itc.2014.6932974</a>
     274<br><br><b>Abstract: </b>We describe the GENI Experiment Engine, a Distributed-Platform-as-a-Service facility designed to be implemented on a distributed testbed or infrastructure. The GEE is intended to provide rapid and convenient access to a distributed infrastructure for simple, easy-to-configure experiments and applications. Specifically, the design goal of the GEE is to permit experimenters and application writers to: (a) allocate a GEE slicelet; (b) deploy a simple experiment or application; (c) run the experiment; (d) collect the results; and (e) tear down the experiment, starting from scratch, within five minutes. The GEE consists of four cooperating services over the GENI infrastructure, which together with pre-allocated slicelets and a pre-allocated network offers a complete, ready to use, sliceable platform over the GENI Infrastructure.
     275</li>
     276<br>
     277
     278
     279
     280<li>
     281<b>Bavier, Andy and Coady, Yvonne and Mack, Tony and Matthews, Chris and Mambretti, Joe and McGeer, Rick and Mueller, Paul and Snoeren, Alex and Yuen, Marco</b>
     282, &quot;GENICloud and transcloud.&quot;
     283Proceedings of the 2012 workshop on Cloud services, federation, and the 8th open cirrus summit, San Jose, California, USA, ACM, New York, NY, USA,
     2842012.
     285doi:10.1145/2378975.2378980.
     286<a href="http://dx.doi.org/10.1145/2378975.2378980">http://dx.doi.org/10.1145/2378975.2378980</a>
     287<br><br><b>Abstract: </b>In this paper, we argue that federation of cloud systems requires a standard API for users to create, manage, and destroy virtual objects, and a standard naming scheme for virtual objects. We introduce an existing API for this purpose, the Slice-Based Federation Architecture, and demonstrate that it can be implemented on a number of existing cloud management systems. We introduce a simple naming scheme for virtual objects, and discuss its implementation.
     288</li>
     289<br>
     290
     291
     292
     293<li>
     294<b>Bejerano, Y. and Ferragut, J. and Guo, K. and Gupta, V. and Gutterman, C. and Nandagopal, T. and Zussman, G.</b>
     295, &quot;Experimental Evaluation of a Scalable WiFi Multicast Scheme in the ORBIT Testbed.&quot;
     296Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     2972014.
     298doi:10.1109/gree.2014.22.
     299<a href="http://dx.doi.org/10.1109/gree.2014.22">http://dx.doi.org/10.1109/gree.2014.22</a>
     300<br><br><b>Abstract: </b>IEEE 802.11-based wireless local area networks, referred to as WiFi, have been globally deployed and the vast majority of the mobile devices are currently WiFi-enabled. While WiFi has been proposed for multimedia content distribution, its lack of adequate support for multicast services hinders its ability to provide multimedia content distribution to a large number of devices. In earlier work, we proposed a dynamic scheme called AMuSe that selects a subset of the multicast receivers as feedback nodes. The feedback nodes periodically send information about channel quality to the multicast sender and the sender in turn can optimize multicast service quality, e.g., by dynamically adjusting transmission bit-rate. In this paper, we discuss several experimental results for the performance evaluation of AMuSe. Our experiments use more than 250 nodes placed in a grid topology in the ORBIT testbed. We consider different experimental scenarios: with and without the presence of external noise. Our focus is on studying the performance of WiFi nodes in WiFi multicast and establishing the conditions that make AMuSe an attractive scheme for feedback in WiFi multicast.
     301</li>
     302<br>
     303
     304
     305
     306<li>
     307<b>Berman, M. and Brinn, M.</b>
     308, &quot;Progress and challenges in worldwide federation of future internet and distributed cloud testbeds.&quot;
     309Science and Technology Conference (Modern Networking Technologies) (MoNeTeC), 2014 First International, IEEE,
     3102014.
     311doi:10.1109/monetec.2014.6995579.
     312<a href="http://dx.doi.org/10.1109/monetec.2014.6995579">http://dx.doi.org/10.1109/monetec.2014.6995579</a>
     313<br><br><b>Abstract: </b>Future Internet and distributed cloud (FIDC) testbeds are rapidly becoming important research and educational resoures worldwide. While FIDC testbeds may be built on diverse technologies, they share the primary capabilities of slicing (virtualized end-to-end configurations of computing, networking, and storage resources) and deep programmability (experimenter programmability of all resources from low level hardware to virtualized components). FIDC testbeds often achieve their deep programmability through software defined networking (SDN) capabilities, which researchers employ both to construct per-application and per-experiment virtual networks, and to intelligently steer traffic throughout the virtual network/cloud environment.
     314</li>
     315<br>
     316
     317
     318
     319<li>
     320<b>Berman, Mark and Chase, Jeffrey S. and Landweber, Lawrence and Nakao, Akihiro and Ott, Max and Raychaudhuri, Dipankar and Ricci, Robert and Seskar, Ivan</b>
     321, &quot;GENI: A federated testbed for innovative network experiments.&quot;
     322Computer Networks,
     3232014.
     324doi:10.1016/j.bjp.2013.12.037.
     325<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.037">http://dx.doi.org/10.1016/j.bjp.2013.12.037</a>
     326<br><br><b>Abstract: </b>GENI, the Global Environment for Networking Innovation, is a distributed virtual laboratory for transformative, at-scale experiments in network science, services, and security. Designed in response to concerns over Internet ossification, GENI is enabling a wide variety of experiments in a range of areas, including clean-slate networking, protocol design and evaluation, distributed service offerings, social network integration, content management, and in-network service deployment. Recently, GENI has been leading an effort to explore the potential of its underlying technologies, SDN and GENI racks, in support of university campus network management and applications. With the concurrent deployment of these technologies on regional and national R&#x0026;E backbones, this will result in a revolutionary new national-scale distributed architecture, bringing to the entire network the shared, deeply programmable environment that the cloud has brought to the datacenter. This deeply programmable environment will support the GENI research mission and as well as enabling research in a wide variety of application areas.
     327</li>
     328<br>
     329
     330
     331
     332<li>
     333<b>Berman, Mark and Demeester, Piet and Lee, Jae W. and Nagaraja, Kiran and Zink, Michael and Colle, Didier and Krishnappa, Dilip K. and Raychaudhuri, Dipankar and Schulzrinne, Henning and Seskar, Ivan and Sharma, Sachin</b>
     334, &quot;Future Internets Escape the Simulator.&quot;
     335Commun. ACM, ACM, New York, NY, USA,
     3362015.
     337doi:10.1145/2699392.
     338<a href="http://dx.doi.org/10.1145/2699392">http://dx.doi.org/10.1145/2699392</a>
     339<br><br><b>Abstract: </b>Future Internet testbeds permit experiments not possible in today's public Net or commercial cloud services.
     340</li>
     341<br>
     342
     343
     344
     345<li>
     346<b>Berman, Mark and Elliott, Chip and Landweber, Lawrence</b>
     347, &quot;GENI: Large-Scale Distributed Infrastructure for Networking and Distributed Systems Research.&quot;
     3482014 IEEE Fifth International Conference on Communications and Electronics (ICCE), Da Nang, Vietnam,
     3492014.
     350doi:10.1109/CCE.2014.6916696.
     351<a href="http://dx.doi.org/10.1109/CCE.2014.6916696">http://dx.doi.org/10.1109/CCE.2014.6916696</a>
     352<br><br><b>Abstract: </b>GENI, the Global Environment for Networking Innovation, is a distributed virtual laboratory for research in networking and distributed systems, with applications in domain science. The main components of GENI include OpenFlow-enabled software defined networking (SDN) resources deployed on over 40 university campuses across the U.S. These resources include both switches and GENI Racks (SDN capable compute clusters with OpenFlow switches for internal and external communications). GENI Racks are currently installed on dozens of university campuses and within R&#x0026;E network backbones. Also available is a diverse group of programmable computing and wireless networking resources. Researchers access this collection of resources via the key GENI techniques of deep programmability and slicing. Collectively, these resources and methods enable GENI to support a wide variety of research efforts.
     353</li>
     354<br>
     355
     356
     357
     358<li>
     359<b>Berryman, Alex and Calyam, Prasad and Cecil, Joe and Adams, George B. and Comer, Douglas</b>
     360, &quot;Advanced Manufacturing Use Cases and Early Results in GENI Infrastructure.&quot;
     3612013 Proceedings Second GENI Research and Educational Experiment Workshop, Salt Lake City, UT, IEEE,
     3622013.
     363doi:10.1109/GREE.2013.13.
     364<a href="http://dx.doi.org/10.1109/GREE.2013.13">http://dx.doi.org/10.1109/GREE.2013.13</a>
     365<br><br><b>Abstract: </b>Providing remote access and collaboration technologies to advanced manufacturing communities are exciting prospects due to the growth of the global marketplace and the pervasiveness of high-speed networks. There is a need to develop reliable protocols that extend beyond the current capabilities of typical TCP/IP connections that do not provide sufficient redundancy for controlling remote processes in manufacturing facilities. In addition, there is a need to suitably configure remote access protocol configurations that deliver satisfactory user experience amongst distributed collaborators synchronously working on manufacturing design workflows using cloud-hosted simulation software. In this paper, we present two case studies and early results that leverage the GENI Future Internet infrastructure for experimentation and development of new services that address such advanced manufacturing needs. Both case studies pivot around the idea of removing the need for users to have physical access to manufacturing resources and thus enable remote access to cloud-hosted services that use Future Internet capabilities for cost/time savings and improved convenience.
     366</li>
     367<br>
     368
     369
     370
     371<li>
     372<b>Bhanage, G. and Daya, R. and Seskar, I. and Raychaudhuri, D.</b>
     373, &quot;VNTS: A Virtual Network Traffic Shaper for Air Time Fairness in 802.16e Systems.&quot;
     374Communications (ICC), 2010 IEEE International Conference on, IEEE,
     3752010.
     376doi:10.1109/ICC.2010.5502484.
     377<a href="http://dx.doi.org/10.1109/ICC.2010.5502484">http://dx.doi.org/10.1109/ICC.2010.5502484</a>
     378<br><br><b>Abstract: </b>The 802.16e standard for broadband wireless access mandates the presence of QoS classes, but does not specify guidelines for the scheduler implementation or mechanisms to ensure air time fairness. Our study demonstrates the feasibility of controlling downlink airtime fairness for slices while running above a proprietary WiMAX basestation (BS) scheduler. We design and implement a virtualized infrastructure that allows users to obtain at least an allocated percentage of BS resources in the presence of saturation and link degradation. Using Kernel virtual machines for creating slices and Click modular router for implementing the virtual network traffic shaping engine we show that it is possible to adaptively control slice usage for downlink traffic on a WiMAX Basestation. The fairness index and coupling coefficient show an improvement of up to 42&#x0025;, and 73&#x0025; with preliminary indoor walking mobility experiments. Outdoor vehicular measurements show an improvement of up to 27&#x0025;, and 70\\\\ with the fairness index and coupling coefficient respectively
     379</li>
     380<br>
     381
     382
     383
     384<li>
     385<b>Bhanage, G. and Vete, D. and Seskar, I. and Raychaudhuri, D.</b>
     386, &quot;SplitAP: Leveraging Wireless Network Virtualization for Flexible Sharing of WLANs.&quot;
     387Global Telecommunications Conference (GLOBECOM 2010), 2010 IEEE, IEEE,
     3882010.
     389doi:10.1109/GLOCOM.2010.5684328.
     390<a href="http://dx.doi.org/10.1109/GLOCOM.2010.5684328">http://dx.doi.org/10.1109/GLOCOM.2010.5684328</a>
     391<br><br><b>Abstract: </b>Providing air-time guarantees across a group of clients forms a fundamental building block in sharing an access point (AP) across different virtual network providers. Though this problem has a relatively simple solution for downlink group scheduling through traffic engineering at the AP, solving this problem for uplink (UL) traffic presents a challenge for fair sharing of wireless hotspots. Among other issues, the mechanism for uplink traffic control has to scale across a large user base, and provide flexible operation irrespective of the client channel conditions and network loads. In this study, we propose the SplitAP architecture that address the problem of sharing uplink airtime across groups of users by extending the idea of network virtualization. Our architecture allows us to deploy different algorithms for enforcing UL airtime fairness across client groups. In this study, we will highlight the design features of the SplitAP architecture, and present results from evaluation on a prototype deployed with: (1) LPFC and (2) LPFC+, two algorithms for controlling UL group fairness. Performance comparisons on the ORBIT testbed show that the proposed algorithms are capable of providing group air-time fairness across wireless clients irrespective of the network volume, and traffic type. The algorithms show up to 40&#x0025; improvement with a modified Jain fairness index.
     392</li>
     393<br>
     394
     395
     396
     397<li>
     398<b>Bhanage, Gautam and Seskar, Ivan and Mahindra, Rajesh and Raychaudhuri, Dipankar</b>
     399, &quot;Virtual basestation: architecture for an open shared WiMAX framework.&quot;
     400Proceedings of the second ACM SIGCOMM workshop on Virtualized infrastructure systems and architectures, New Delhi, India, ACM, New York, NY, USA,
     4012010.
     402doi:10.1145/1851399.1851401.
     403<a href="http://dx.doi.org/10.1145/1851399.1851401">http://dx.doi.org/10.1145/1851399.1851401</a>
     404<br><br><b>Abstract: </b>This paper presents the architecture and performance evaluation of a virtualized wide-area &#x34;&#x0308;G&#x20;&#x0308;cellular wireless network. Specifically, it addresses the challenges of virtualization of resources in a cellular base station to enable shared use by multiple independent slice users (experimenters or mobile virtual network operators), each with possibly distinct flow types and network layer protocols. The proposed virtual basestation architecture is based on an external substrate which uses a layer-2 switched datapath, and an arbitrated control path to the WiMAX basestation. The framework implements virtualization of base station's radio resources to achieve isolation between multiple virtual networks. An algorithm for weighted fair sharing among multiple slices based on an airtime fairness metric has been implemented for the first release. Preliminary experimental results from the virtual basestation prototype are given, demonstrating mobile network performance, isolation across slices with different flow types, and custom flow scheduling capabilities.
     405</li>
     406<br>
     407
     408
     409
     410<li>
     411<b>Bhanage, Gautam and Seskar, Ivan and Raychaudhuri, Dipankar</b>
     412, &quot;A virtualization architecture for mobile WiMAX networks.&quot;
     413SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
     4142012.
     415doi:10.1145/2169077.2169082.
     416<a href="http://dx.doi.org/10.1145/2169077.2169082">http://dx.doi.org/10.1145/2169077.2169082</a>
     417<br><br><b>Abstract: </b>Systems virtualization offers convenient means for sharing networking infrastructure while improving its utilization. This study addresses the challenges of virtualizing a commercial off-the-shelf 4G mobileWiMAX basestation. We highlight additions and modifications needed in theWiMAX network architecture for supporting multiple simultaneous virtual basestations on a single physical basestation. The most prominent features provided by the proposed virtual basestation framework include the capability to perform all frame switching at layer-2, and control mechanisms to provide isolation across slices needed to ensure experiment repeatability. By prototyping on a commercial WiMAX radio, this paper shows the usage of the virtual basestation system for housing mobile virtual network operators and testbeds alike. A use case is shown where the virtual basestation design is used to evaluate mobile handoff schemes. Another usage case is shown for optimizing a video delivery on the edge. The video delivery use case is used to show performance improvements of up to 5dB in the PSNR. Evaluation of prototype shows a significant improvement in the slice isolation, with aggregate throughput improvements of up to 192&#x0025; achievable through fair resource allocation.
     418</li>
     419<br>
     420
     421
     422
     423<li>
     424<b>Bhanage, Gautam and Seskar, Ivan and Zhang, Yanyong and Raychaudhuri, Dipankar and Jain, Shweta</b>
     425, &quot;Experimental Evaluation of OpenVZ from a Testbed Deployment Perspective.&quot;
     426Testbeds and Research Infrastructures. Development of Networks and Communities, Springer Berlin Heidelberg,
     4272011.
     428doi:10.1007/978-3-642-17851-1&#x005F;7.
     429<a href="http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;7">http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;7</a>
     430<br><br><b>Abstract: </b>A scalable approach to building large scale experimentation testbeds involves multiplexing the system resources for better utilization. Virtualization provides a convenient means of sharing testbed resources among experimenters. The degree of programmability and isolation achieved with such a setup is largely dependent on the type of technology used for virtualization. We consider OpenVZ and User Mode Linux (UML) for virtualization of the ORBIT wireless testbed and evaluate their relative merit. Our results show that OpenVZ, an operating system level virtualization mechanism significantly outperforms UML in terms of system overheads and performance isolation. We discuss both qualitative and quantitative performance features which could serve as guidelines for selection of a virtualization scheme for similar testbeds.
     431</li>
     432<br>
     433
     434
     435
     436<li>
     437<b>Bhat, Divyashri and Riga, Niky and Zink, Michael</b>
     438, &quot;Towards seamless application delivery using software defined exchanges.&quot;
     439Teletraffic Congress (ITC), 2014 26th International, IEEE,
     4402014.
     441doi:10.1109/itc.2014.6932971.
     442<a href="http://dx.doi.org/10.1109/itc.2014.6932971">http://dx.doi.org/10.1109/itc.2014.6932971</a>
     443<br><br><b>Abstract: </b>Content Delivery over the Internet continues to be a challenge as there is no centralized control system [1]. Software Defined Networking has paved the way to provide this control of network traffic. OpenFlow is now being standardized as part of the Open Networking Foundation, and Software Defined Exchange provides a framework to use OpenFlow for multidomain routing. Prototype deployments of Software Defined Exchanges have recently come into existence as a platform for Future Internet architecture to eliminate the need for core routing technology used in today's Internet. In this paper, we look at how application delivery, in particular, Dynamic Adaptive Streaming over HTTP (DASH) and Nowcasting take advantage of Software Defined Exchange. We compare unsophisticated controllers to more sophisticated ones which we call a ” load balancer” and find that implementing a good reactive controller for inter-domain routing can result in better network utilization and better application performance.
     444</li>
     445<br>
     446
     447
     448
     449<li>
     450<b>Bhat, Divyashri and Wang, Cong and Rizk, Amr and Zink, Michael</b>
     451, &quot;A load balancing approach for adaptive bitrate streaming in Information Centric networks.&quot;
     452Multimedia &#x0026; Expo Workshops (ICMEW), 2015 IEEE International Conference on, IEEE,
     4532015.
     454doi:10.1109/icmew.2015.7169802.
     455<a href="http://dx.doi.org/10.1109/icmew.2015.7169802">http://dx.doi.org/10.1109/icmew.2015.7169802</a>
     456<br><br><b>Abstract: </b>The Information Centric Networking (ICN) paradigm promises deconstraining the current Internet architecture by allowing clients to directly address the desired content throughout the network. For the Internet this is a further evolutionary step from the idea of a narrow-waist core that only transports requests/replies to an intelligent architecture searching for and providing content. Multi-sourcing, which is one of the core ideas of ICN, constitutes a serious challenge for prevalent Internet applications such as video streaming. In this work we show how prominent adaptive video streaming protocols can benefit from the load balancing capabilities that are native to ICN. We examine the performance of content retrieval in ICN over Ethernet in a real-world testbed showing the impact of multi-sourcing and content size variation on the content transfer times.
     457</li>
     458<br>
     459
     460
     461
     462<li>
     463<b>Bhojwani, Sushil</b>
     464, &quot;Interoperability in Federated Clouds (Master's thesis).&quot;
     465
     4662015.
     467
     468<a href="http://hdl.handle.net/1828/6732">http://hdl.handle.net/1828/6732</a>
     469<br><br><b>Abstract: </b>Cloud Computing is the new trend in sharing resources, sharing and managing data and performing computations on a shared resource via the Internet. However, with the constant increase in demand, these resources are insufficient. So users often use another network in conjunction with the current one. All these networks accomplish the goal of providing the user with a virtual or physical machine. However, to achieve the result, virtual machine users have to maintain multitude credentials and follow a different process for each network. In this thesis, we focus on SAGEFed, a product that enables a user to use the same credentials and commands to reserve the resources on two different federated clouds, i.e., SAVI and GENI. As a part of SAGEFed, the user can acquire or reserve resources on the clouds with the same API. The same service also manages the credentials, so they do not have to manage different credentials while acquiring resources. Furthermore, SAGEFed demonstrates that any cloud that has some form of client tool can be easily integrated.
     470</li>
     471<br>
     472
     473
     474
     475<li>
     476<b>Bhojwani, Sushil and Hemmings, Matt and Ingalls, Dan and Lincke, Jens and Krahn, Robert and Lary, David and McGeer, Rick and Ricart, Glenn and Roder, Marko and Coady, Yvonne and Stege, Ulrike</b>
     477, &quot;The Ignite Distributed Collaborative Visualization System.&quot;
     478SIGMETRICS Perform. Eval. Rev., ACM, New York, NY, USA,
     4792015.
     480doi:10.1145/2847220.2847234.
     481<a href="http://dx.doi.org/10.1145/2847220.2847234">http://dx.doi.org/10.1145/2847220.2847234</a>
     482<br><br><b>Abstract: </b>An abstract is not available.
     483</li>
     484<br>
     485
     486
     487
     488<li>
     489<b>Blanton, Ethan and Chatterjee, Sarbajit and Gangam, Sriharsha and Kala, Sumit and Sharma, Deepti and Fahmy, Sonia and Sharma, Puneet</b>
     490, &quot;Design and evaluation of the S<sup>3</sup> monitor network measurement service on GENI.&quot;
     4912012 Fourth International Conference on Communication Systems and Networks (COMSNETS 2012), Bangalore, India, IEEE,
     4922012.
     493doi:10.1109/COMSNETS.2012.6151327.
     494<a href="http://dx.doi.org/10.1109/COMSNETS.2012.6151327">http://dx.doi.org/10.1109/COMSNETS.2012.6151327</a>
     495<br><br><b>Abstract: </b>Network monitoring capabilities are critical for both network operators and networked applications. In the context of an experimental test facility, network measurement is important for researchers experimenting with new network architectures and applications, as well as operators of the test facility itself. The Global Environment for Network Innovations (GENI) is a sophisticated test facility comprised of multiple ” control frameworks.” In this paper, we describe the design and implementation of S
     496</li>
     497<br>
     498
     499
     500
     501<li>
     502<b>Brinn, Marshall and Bastin, NIcholas and Bavier, Andrew and Berman, Mark and Chase, Jeffrey and Ricci, Robert</b>
     503, &quot;Trust as the Foundation of Resource Exchange in GENI.&quot;
     504Proceedings of the 10th EAI International Conference on Testbeds and Research Infrastructures for the Development of Networks &#x0026; Communities, Vancouver, Canada, ACM,
     5052015.
     506doi:10.4108/icst.tridentcom.2015.259683.
     507<a href="http://dx.doi.org/10.4108/icst.tridentcom.2015.259683">http://dx.doi.org/10.4108/icst.tridentcom.2015.259683</a>
     508<br><br><b>Abstract: </b>Researchers and educators in computer science and other domains are increasingly turning to distributed test beds that offer access to a variety of resources, including networking, computation, storage, sensing, and actuation. The provisioning of resources from their owners to interested experimenters requires establishing sufficient mutual trust between these parties. Building such trust directly between researchers and resource owners will not scale as the number of experimenters and resource owners grows. The NSF GENI (Global Environment for Network Innovation) project has focused on establishing scalable mechanisms for maintaining such trust based on common approaches for authentication, authorization and accountability. Such trust reflects the actual trust relationships and agreements among humans or real-world organizations. We describe here GENI's approaches for federated trust based on mutually trusted authorities, and implemented via cryptographically signed credentials and shared policies.
     509</li>
     510<br>
     511
     512
     513
     514<li>
     515<b>Bronzino, Francesco and Han, Chao and Chen, Yang and Nagaraja, Kiran and Yang, Xiaowei and Seskar, Ivan and Raychaudhuri, Dipankar</b>
     516, &quot;In-Network Compute Extensions for Rate-Adaptive Content Delivery in Mobile Networks.&quot;
     517Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     5182014.
     519doi:10.1109/icnp.2014.81.
     520<a href="http://dx.doi.org/10.1109/icnp.2014.81">http://dx.doi.org/10.1109/icnp.2014.81</a>
     521<br><br><b>Abstract: </b>Traffic from mobile wireless networks has been growing at a fast pace in recent years and is expected to surpass wired traffic very soon. Service providers face significant challenges at such scales including providing seamless mobility, efficient data delivery, security, and provisioning capacity at the wireless edge. In the Mobility First project, we have been exploring clean slate enhancements to the network protocols that can inherently provide support for at-scale mobility and trustworthiness in the Internet. An extensible data plane using pluggable compute-layer services is a key component of this architecture. We believe these extensions can be used to implement in-network services to enhance mobile end-user experience by either off-loading work and/or traffic from mobile devices, or by enabling en-route service-adaptation through context-awareness (e.g., Knowing contemporary access bandwidth). In this work we present details of the architectural support for in-network services within Mobility First, and propose protocol and service-API extensions to flexibly address these pluggable services from end-points. As a demonstrative example, we implement an in network service that does rate adaptation when delivering video streams to mobile devices that experience variable connection quality. We present details of our deployment and evaluation of the non-IP protocols along with compute-layer extensions on the GENI test bed, where we used a set of programmable nodes across 7 distributed sites to configure a Mobility First network with hosts, routers, and in-network compute services.
     522</li>
     523<br>
     524
     525
     526
     527<li>
     528<b>Brown, D. and Ascigil, O. and Nasir, H. and Carpenter, C. and Griffioen, J. and Calvert, K.</b>
     529, &quot;Designing a GENI Experimenter Tool to Support the Choice Net Internet Architecture.&quot;
     530Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     5312014.
     532doi:10.1109/icnp.2014.88.
     533<a href="http://dx.doi.org/10.1109/icnp.2014.88">http://dx.doi.org/10.1109/icnp.2014.88</a>
     534<br><br><b>Abstract: </b>Test beds such as GENI provide an ideal environment for experimenting with future internet architectures such as Choice Net. Unlike the narrow waist of the current Internet (IP), Choice Net encourages alternatives and competition at the network layer via an economic plane that allows users to choose and purchase precisely the services they need. In this paper we describe our experiences implementing the Choice Net architecture on GENI. Some features of GENI, such as the ability to program the network layer, to leverage existing protocols and software, to run real applications generating realistic traffic, and the ability to perform long-running experiments made GENI an ideal platform for Choice Net experimentation. However, we found that GENI currently lacks the tools needed to make it easy to use these features. To address this issue, we designed and implemented a GENI Experimenter Tool specifically designed and tailored to perform tasks commonly needed by experimenters such as dynamically configuring nodes, loading and compiling node-specific code, executing Click modules, running commands on sets of nodes, accessing the local file system on nodes, and dynamically logging into nodes.
     535</li>
     536<br>
     537
     538
     539
     540<li>
     541<b>Brown, D. and Nasir, H. and Carpenter, C. and Ascigil, O. and Griffioen, J. and Calvert, K.</b>
     542, &quot;ChoiceNet gaming: Changing the gaming experience with economics.&quot;
     543Computer Games: AI, Animation, Mobile, Multimedia, Educational and Serious Games (CGAMES), 2014, IEEE,
     5442014.
     545doi:10.1109/cgames.2014.6934146.
     546<a href="http://dx.doi.org/10.1109/cgames.2014.6934146">http://dx.doi.org/10.1109/cgames.2014.6934146</a>
     547<br><br><b>Abstract: </b>When playing online games, the user experience is often dictated by the performance of the network. To deliver the best possible gaming experience, game developers often find themselves developing work-arounds that try to mask the lack of control they have over of the existing TCP/IP Internet. ChoiceNet, an emerging future Internet architecture, attempts to give applications enhanced control (choice) over the service they receive from the network. In particular, ChoiceNet supports an economic plane in which applications can purchase services from any provider. Because providers are compensated, they are motivated to offer a variety of innovative, excellent services, enabling applications to select the service best suited for its needs. Instead of coding work-arounds, game developers can obtain precisely the network service that is needed to optimize the game experience. In this paper, we describe the emerging ChoiceNet archi- tecture and show how computer games can benefit from the alternatives enabled by ChoiceNet. To demonstrate the benefits of the ChoiceNet architecture, we implemented a first person shooter game that uses ChoiceNet to ” purchase” and then send data over the purchased path resulting in substantially lower latency than the default path. We describe the ChoiceNet services used to implement the game, and we present performance results that show a significant reduction in latency. We also show how ChoiceNet can be used to purchase reliable (non-lossy) communication paths that improve the user's experience.
     548</li>
     549<br>
     550
     551
     552
     553<li>
     554<b>Calyam, P. and Rajagopalan, S. and Selvadhurai, A. and Mohan, S. and Venkataraman, A. and Berryman, A. and Ramnath, R.</b>
     555, &quot;Leveraging OpenFlow for resource placement of virtual desktop cloud applications.&quot;
     556Integrated Network Management (IM 2013), 2013 IFIP/IEEE International Symposium on,
     5572013.
     558
     559
     560<br><br><b>Abstract: </b>Popular applications such as email, photo/video galleries, and file storage are increasingly being supported by cloud platforms in residential, academia and industry communities. The next frontier for these user communities will be to transition `traditional desktops' that have dedicated hardware and software configurations into `virtual desktop clouds' that are accessible via thin-clients. In this paper, we describe an intelligent resource placement framework for thin-client based virtual desktops. The framework leverages principles of softwaredefined networking and features a `unified resource broker' that uses special `marker packets' for: (a) ” route setup” when handling non-IP traffic between thin-client sites and data centers, (b) ” path selection” and ” load balancing” of virtual desktop flows to improve performance of interactive applications and video playback, and to cope with faults such as link-failures or Denialof-Service cyber-attacks. In addition, we detail our framework implementation within a virtual desktop cloud (VDC) setup in a multi-domain Global Environment for Network Innovations (GENI) Future Internet testbed spanning backbone and access networks. We present empirical results from our experimentation that leverages OpenFlow programmable networking, as well as perfSONAR instrumentation-and-measurement capabilities for validating our framework in GENI under realistic settings. Our results demonstrate the importance of scheduling regulated measurements that can be used for intelligent resource placement decisions. Our results also show the feasibility and benefits of using OpenFlow controller applications for path selection and load balancing between thin-client sites and data centers in VDCs.
     561</li>
     562<br>
     563
     564
     565
     566<li>
     567<b>Calyam, P. and Seetharam, S. and Antequera, R. B.</b>
     568, &quot;GENI Laboratory Exercises Development for a Cloud Computing Course.&quot;
     569Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     5702014.
     571doi:10.1109/gree.2014.15.
     572<a href="http://dx.doi.org/10.1109/gree.2014.15">http://dx.doi.org/10.1109/gree.2014.15</a>
     573<br><br><b>Abstract: </b>Cloud computing education involves integration of computing theories and information technologies in new and interesting ways. It can enable students to architect scalable infrastructures and develop web-service based applications utilizing distributed systems. In this paper, we describe our efforts, experiences and findings in the development of laboratory exercises that utilize GENI infrastructure in a cloud computing course offered at University of Missouri in Fall 2013. Three sets of laboratory exercises were developed and administered for 30 undergraduate/graduate students to help them gain skills in computer and network virtualization, and also to prepare them for distributed system programming projects. We found that the GENI infrastructure provides unique capabilities for student training, and combining it with lab exercises that use public clouds such as Amazon Web Services can provide an overall rich set of hands-on learning opportunities.
     574</li>
     575<br>
     576
     577
     578
     579<li>
     580<b>Calyam, P. and Sridharan, M. and Xu, Yingxiao and Zhu, Kunpeng and Berryman, A. and Patali, R. and Venkataraman, A.</b>
     581, &quot;Enabling performance intelligence for application adaptation in the Future Internet.&quot;
     582Communications and Networks, Journal of,
     5832011.
     584doi:10.1109/JCN.2011.6157475.
     585<a href="http://dx.doi.org/10.1109/JCN.2011.6157475">http://dx.doi.org/10.1109/JCN.2011.6157475</a>
     586<br><br><b>Abstract: </b>Today's Internet which provides communication channels with best-effort end-to-end performance is rapidly evolving into an autonomic global computing platform. Achieving autonomicity in the Future Internet will require a performance architecture that (a) allows users to request and own 'slices' of geographically-distributed host and network resources, (b) measures and monitors end-to-end host and network status, (c) enables analysis of the measurements within expert systems, and (d) provides performance intelligence in a timely manner for application adaptations to improve performance and scalability. We de- scribe the requirements and design of one such &#x46;&#x0308;uture Internet performance architecture&#x20;&#x0308;(FIPA), and present our reference implementation of FIPA called 'OnTimeMeasure.' OnTimeMeasure comprises of several measurement-related services that can interact with each other and with existing measurement frameworks to enable performance intelligence. We also explain our OnTimeMea- sure deployment in the global environment for network innovations (GENI) infrastructure collaborative research initiative to build a sliceable Future Internet. Further, we present an application- adaptation case study in GENI that uses OnTimeMeasure-enabled performance intelligence in the context of dynamic resource allocation within thin-client based virtual desktop clouds. We show how a virtual desktop cloud provider in the Future Internet can use the performance intelligence to increase cloud scalability, while simultaneously delivering satisfactory user quality-of-experience.
     587</li>
     588<br>
     589
     590
     591
     592<li>
     593<b>Calyam, Prasad and Mishra, Anup and Antequera, Ronny B. and Chemodanov, Dmitrii and Berryman, Alex and Zhu, Kunpeng and Abbott, Carmen and Skubic, Marjorie</b>
     594, &quot;Synchronous Big Data analytics for personalized and remote physical therapy.&quot;
     595Pervasive and Mobile Computing,
     5962015.
     597doi:10.1016/j.pmcj.2015.09.004.
     598<a href="http://dx.doi.org/10.1016/j.pmcj.2015.09.004">http://dx.doi.org/10.1016/j.pmcj.2015.09.004</a>
     599<br><br><b>Abstract: </b>With gigabit networking becoming economically feasible and widely installed at homes, there are new opportunities to revisit in-home, personalized telehealth services. In this paper, we describe a novel telehealth eldercare service that we developed viz., ” PhysicalTherapy-as-a-Service” (PTaaS) that connects a remote physical therapist at a clinic to a senior at home. The service leverages a high-speed, low-latency network connection through an interactive interface built on top of Microsoft Kinect motion sensing capabilities. The interface that is built using user-centered design principles for wellness coaching exercises is essentially a 'Synchronous Big Data' application due to its: (i) high data-in-motion velocity (i.e., peak data rate is ≈400 Mbps), (ii) considerable variety (i.e., measurements include 3D sensing, network health, user opinion surveys and video clips of RGB, skeletal and depth data), and (iii) large volume (i.e., several GB of measurement data for a simple exercise activity). The successful PTaaS delivery through this interface is dependent on the veracity analytics needed for correlation of the real-time Big Data streams within a session, in order to assess exercise balance of the senior without any bias due to network quality effects. Our experiments with PTaaS in an actual testbed involving senior homes in Kansas City with Google Fiber connections and our university clinic demonstrate the network configuration and time synchronization related challenges in order to perform online analytics. Our findings provide insights on how to: (a) enable suitable resource calibration and perform network troubleshooting for high user experience for both the therapist and the senior, and (b) realize a Big Data architecture for PTaaS and other similar personalized healthcare services to be remotely delivered at a large-scale in a reliable, secure and cost-effective manner.
     600</li>
     601<br>
     602
     603
     604
     605<li>
     606<b>Calyam, Prasad and Rajagopalan, Sudharsan and Seetharam, Sripriya and Selvadhurai, Arunprasath and Salah, Khaled and Ramnath, Rajiv</b>
     607, &quot;VDC-Analyst: Design and verification of virtual desktop cloud resource allocations.&quot;
     608Computer Networks,
     6092014.
     610doi:10.1016/j.comnet.2014.02.022.
     611<a href="http://dx.doi.org/10.1016/j.comnet.2014.02.022">http://dx.doi.org/10.1016/j.comnet.2014.02.022</a>
     612<br><br><b>Abstract: </b>One of the significant challenges for Cloud Service Providers (CSPs) hosting ” virtual desktop cloud” (VDC) infrastructures is to deliver a satisfactory quality of experience (QoE) to the user. In order to maximize the user QoE without expensive resource overprovisioning, there is a need to design and verify resource allocation schemes for a comprehensive set of VDC configurations. In this paper, we present ” VDC-Analyst”, a novel tool that can capture critical quality metrics such as Net Utility and Service Response Time, which can be used to quantify VDC platform readiness. This tool allows CSPs, researchers and educators to design and verify various resource allocation schemes using both simulation and emulation in two modes: ” Run Simulation” and ” Run Experiment”, respectively. The Run Simulation mode allows users to test and visualize resource provisioning and placement schemes on a simulation framework. Run Experiment mode allows testing on a real software-defined network testbed using emulated virtual desktop application traffic to create a realistic environment. Results from using our tool demonstrate that a significant increase in perceived user QoE can be achieved by using a combination of the following techniques incorporated in the tool: (i) optimizing Net Utility through a ” Cost-Aware Utility-Maximal Resource Allocation Algorithm”, (ii) estimating values for Service Response Time using a ” Multi-stage Queuing Model”, and (iii) appropriate load balancing through software-defined networking adaptations in the VDC testbed.
     613</li>
     614<br>
     615
     616
     617
     618<li>
     619<b>Calyam, Prasad and Venkataraman, Aishwarya and Berryman, Alex and Faerman, Marcio</b>
     620, &quot;Experiences from Virtual Desktop CloudExperiments in GENI.&quot;
     621First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     6222012.
     623
     624
     625<br><br><b>Abstract: </b>Popular applications such as email, photo/video galleries, and file storage are increasingly being supported by cloud platforms in residential, academia and industry communities. The next frontier for these user communities will be to transition 'traditional desktops' that have dedicated hardware and software configurations into 'virtual desktop clouds' that are accessible via thin-clients. In this paper, we describe experiences from our research and development of virtual desktop cloud experiments in GENI. Our experimentation goal is to investigate and develop optimal resource allocation frameworks and performance bench- marking tools that can enable provisioning (i.e., resource sizing) and placement (i.e., resource mapping) of thin-client based virtual desktops at Internet-scale. We first motivate why virtual desktop cloud experiments cannot be done only at a table-top level, and why infrastructures such as GENI are essential. Next, we detail the methodology of our completed ” provisioning” experiments, and our work-in-progress ” placement” experiments in GENI that leverage multiple kinds of GENI resources such as aggregates, measurement services and experimenter workflow tools, as well as commercial software. Lastly, we present our vision on how our experiment slice setup and application development experiences, as well as outcomes can be leveraged in classroom labs, and 'living labs' that use GENI resources to foster training and wide- adoption of Future Internet applications.
     626</li>
     627<br>
     628
     629
     630
     631<li>
     632<b>Cameron, Katherine and Brooks, R. R. and Deng, Juan and Yu, Lu and Wang, K. C. and Martin, James</b>
     633, &quot;WiMAX: Bandwidth Contention Resolution Vulnerability to Denial of Service Attacks.&quot;
     634First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     6352012.
     636
     637
     638<br><br><b>Abstract: </b>Wireless communications is part of everyday life and 4G technology, including WiMAX, offers higher data rates and wider coverage than predecessor 3G technologies. Many security vulnerabilities have been discovered in 3G protocols and these vulnerabilities may still exist in next generation 4G protocols. This paper examines how system parameters for the WiMAX Bandwidth Contention Resolution process can affect network vulnerability to DoS attacks. It will present software simulations that explore system parameter settings and will cover the current phase of hardware simulations.
     639</li>
     640<br>
     641
     642
     643
     644<li>
     645<b>Chakrabortty, Aranya and Xin, Yufeng</b>
     646, &quot;Hardware-in-the-Loop Simulations and Verifications of Smart Power Systems Over an Exo-GENI Testbed.&quot;
     6472013 Proceedings Second GENI Research and Educational Experiment Workshop, Salt Lake City, UT, IEEE,
     6482013.
     649doi:10.1109/GREE.2013.12.
     650<a href="http://dx.doi.org/10.1109/GREE.2013.12">http://dx.doi.org/10.1109/GREE.2013.12</a>
     651<br><br><b>Abstract: </b>In this paper we describe an advanced hardware-in- loop simulation facility for real-time demonstration and validation of power system monitoring and control algorithms, currently under construction at NC State University. This facility integrates a real-time power system emulation lab with the GENI network and its associated cloud testbeds. The dynamic responses from the power system emulator are captured via real hardware Phasor Measurement Units (PMU) that are synchronized with the time-scale of the simulations via a common GPS reference. These responses are then sent to the computing and storage resource in GENI using the IEEE C37.118 protocol, running the smart grid control and management application simulations via QoS-guaranteed communications channels, all provisioned in a dynamic fashion.
     652</li>
     653<br>
     654
     655
     656
     657<li>
     658<b>Chen, Kang and Shen, Haiying</b>
     659, &quot;Global optimization of file availability through replication for efficient file sharing in MANETs.&quot;
     660Network Protocols (ICNP), 2011 19th IEEE International Conference on, Vancouver, AB, Canada, IEEE,
     6612011.
     662doi:10.1109/icnp.2011.6089056.
     663<a href="http://dx.doi.org/10.1109/icnp.2011.6089056">http://dx.doi.org/10.1109/icnp.2011.6089056</a>
     664<br><br><b>Abstract: </b>File sharing applications in mobile ad hoc networks (MANETs) have attracted more and more attention in recent years. The efficiency of file querying suffers from the distinctive properties of MANETs including node mobility and limited communication range and resource. An intuitive method to alleviate this problem is to create file replicas in the network. However, despite the efforts on file replication, no research has focused on the global optimal replica sharing with minimum average querying delay. Specifically, current file replication protocols in MANETs have two shortcomings. First, they lack a rule to allocate limited resource to different files in order to minimize the average querying delay. Second, they simply consider storage as resource for replicas, but neglect the fact that the file holders' frequency of meeting other nodes also plays an important role in determining file availability. A node having a higher meeting frequency with others provides higher availability to its files. In this paper, we introduce a new concept of resource for file replication, which considers both node storage and meeting frequency. We theoretically study the influence of resource allocation on the average querying delay and derive a resource allocation rule to minimize the average querying delay. We further propose a distributed file replication protocol that follows the rule. The trace-driven experiments on both the real-world GENI testbed and NS-2 show that our protocol can achieve shorter average querying delay at lower cost than current replication protocols, which justifies the correctness of our theoretical analysis and the effectiveness of the proposed protocol.
     665</li>
     666<br>
     667
     668<li>
     669<b>Chen, Kang and Shen, Haiying</b>
     670, &quot;Cont2: Social-Aware Content and Contact Based File Search in Delay Tolerant Networks.&quot;
     671Proceedings of the 2013 42Nd International Conference on Parallel Processing, IEEE Computer Society, Washington, DC, USA,
     6722013.
     673doi:10.1109/icpp.2013.28.
     674<a href="http://dx.doi.org/10.1109/icpp.2013.28">http://dx.doi.org/10.1109/icpp.2013.28</a>
     675<br><br><b>Abstract: </b>In this paper, we focus on distributed file search over a delay tolerant network (DTN) formed by mobile devices that exhibit the characteristics of social networks. Current file search methods in MANETs/DTNs are either content-based or contact-based. The former builds routing tables for node contents but is not resilient to high node mobility, while the latter exploits node contact patterns in the social networks but may lead to high latency. Recent research also reveal the importance of interests in realizing efficient file dissemination in DTNs. In this paper, we first analyze node interest and mobility from real traces, which confirms the shortcomings of a contact based method and show the importance of considering both content/interest and contact in file search. We then propose Cont2, a social-aware file search method which leverages both node social interests (content) and contact patterns to enhance search efficiency. First, considering people with common interests tend to share files and gather together, Cont2 virtually groups common-interest nodes into a community to direct file search. Second, considering human mobility follows a certain pattern, Cont2 exploits nodes that have high contact frequency with the queried content. Third, Cont2 also exploits active nodes that have more connections to others as a complementary approach to expedite file search. Trace-driven experimental on the real-world GENI test bed and NS-2 simulator show that Cont2 can significantly improve the search efficiency compared to current methods.
     676</li>
     677<br>
     678
     679
     680
     681<li>
     682<b>Chen, Kang and Shen, Haiying and Zhang, Haibo</b>
     683, &quot;Leveraging Social Networks for P2P Content-Based File Sharing in Mobile Ad Hoc Networks.&quot;
     6842011 IEEE Eighth International Conference on Mobile Ad-Hoc and Sensor Systems, Valencia, Spain, IEEE,
     6852011.
     686doi:10.1109/MASS.2011.24.
     687<a href="http://dx.doi.org/10.1109/MASS.2011.24">http://dx.doi.org/10.1109/MASS.2011.24</a>
     688<br><br><b>Abstract: </b>Current P2P file sharing methods in mobile ad hoc networks (MANETs) can be classified into three groups: flooding-based, advertisement-based and social contact-based. The first two groups of methods can easily generate high overhead and low scalability, and the third group fails to consider the social interests (content) of mobile nodes, which otherwise can improve file searching efficiency. In this paper, we propose a P2P content-based file sharing system for MANETs. The system uses an interest extraction algorithm to derive a node's interests from its files for complex queries. For efficient file searching, it groups common-interest nodes that frequently meet with each other as communities. Further, it takes advantage of node mobility by designating stable nodes, which has frequent contact with community members, as community coordinators for intra-community searching, and highly-mobile nodes as community ambassadors for inter-community searching. An interest-oriented file searching scheme further enhances the file searching success rate. We first deployed our system on the real-world GENI Orbit testbed with a real trace and then conducted experiment on the ns2 simulator with both real trace and simulated disconnected and connected MANET scenario. The test results show that our system significantly lowers transmission cost and improves file searching success rate compared to current methods.
     689</li>
     690<br>
     691
     692
     693
     694<li>
     695<b>Chen, Kang and Xu, Ke and Winburn, Steven and Shen, Haiying and Wang, Kuang-Ching and Li, Ze</b>
     696, &quot;Experimentation of a MANET Routing Algorithm on the GENI ORBIT Testbed.&quot;
     697First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     6982012.
     699
     700
     701<br><br><b>Abstract: </b>This paper proposes a systematic procedure for experimentation of Mobile ad hoc networks (MANETs) on the ORBIT testbed. MANETs have attracted significant re- search interests in recent years. Most of routing or file sharing algorithms in MANETs were only evaluated by theoretical analysis or simulations because of the requirement of large scale networks. However, due to the distinctive properties of MANETs, such as mobility and decentralized structure, it has been non-trivial to deploy a real testbed for the verification. The Global Environment for Network Innovations (GENI) project sponsored by the National Science Foundation (NSF) provides an exploratory environment for academic real-world experiments, such as the ORBIT testbed. A stable and repeatable procedure for experimentation on real testbeds is necessary and important to assure the validity of results. In this paper, a MANET routing algorithm, namely LORD, was tested on the ORBIT testbed, using the proposed procedure. Specifically, we first configure the wireless interface on each node to enable the communication between each pair of nodes. Then a set of methods are adopted to construct the MANETs scenario for test. The network status is monitored throughout the entire duration of experiments. Finally, the experiment results of LORD on the GENI ORBIT testbed are demonstrated.
     702</li>
     703<br>
     704
     705
     706
     707<li>
     708<b>Chen, X. and Cai, H. and Wolf, T.</b>
     709, &quot;Multi-criteria Routing in Networks with Path Choices.&quot;
     7102015 IEEE 23rd International Conference on Network Protocols (ICNP), IEEE,
     7112015.
     712doi:10.1109/icnp.2015.36.
     713<a href="http://dx.doi.org/10.1109/icnp.2015.36">http://dx.doi.org/10.1109/icnp.2015.36</a>
     714<br><br><b>Abstract: </b>Typical routing algorithms use a single criterion, such as hop count or link weight, to calculate paths. As the requirement of flexible routing arises, there are circumstances where multiple criteria are needed for routing. Though there are proposed solutions to the multi-criteria optimal path selection problem for quality-of-service routing, they usually combine all criteria into a single path optimization metric a priori. However, this approach is not feasible in scenarios where the path consumers' weightings of criteria is not known at compute time. Such circumstances require finding all the Pareto-optimal paths, i.e., all the paths that are not dominated by other paths. In this paper, we present the algorithmic foundations for efficiently computing Pareto-optimal paths. We present ParetoBFS, a variant of a breadth-first search that uses branch-and-bound techniques to find all the Pareto-optimal paths while effectively limiting the potentially very large search space. We present several sampling techniques to further increase the speed of the search while degrading the quality of the results only marginally. Our simulation results show that existing multi-criteria combinatorial optimization approaches can only search a small fraction of all the Pareto-optimal paths while ParetoBFS can obtain the whole path set in shorter time. We also present results from an implementation of ParetoBFS on a software-defined network prototype.
     715</li>
     716<br>
     717
     718
     719
     720<li>
     721<b>Chen, Xinming and Wolf, Tilman and Griffioen, Jim and Ascigil, Onur and Dutta, Rudra and Rouskas, George and Bhat, Shireesh and Baldin, Ilya and Calvert, Ken</b>
     722, &quot;Design of a protocol to enable economic transactions for network services.&quot;
     723Communications (ICC), 2015 IEEE International Conference on, IEEE,
     7242015.
     725doi:10.1109/icc.2015.7249175.
     726<a href="http://dx.doi.org/10.1109/icc.2015.7249175">http://dx.doi.org/10.1109/icc.2015.7249175</a>
     727<br><br><b>Abstract: </b>Deployment of innovative new networking services requires support by network providers. Since economic motivation plays an important role for network providers, it is critical that a network architecture intrinsically considers economic relationships. We present the design of a protocol that associates access to network services with economic contracts. We show how this protocol can be realized in fundamentally different ways, using out-of-band signaling and in-band signaling, based on two different prototype implementations. We present results that show the effectiveness of the proposed protocol and thus demonstrate a first step toward realizing an economy plane for the Internet.
     728</li>
     729<br>
     730
     731
     732
     733<li>
     734<b>Cherukuri, Ramkumar and Liu, Xuan and Bavier, Andy and Sterbenz, James P. G. and Medhi, Deep</b>
     735, &quot;Network virtualization in GpENI: Framework, implementation &#x0026;amp; integration experience.&quot;
     73612th IFIP/IEEE International Symposium on Integrated Network Management (IM 2011) and Workshops, Dublin, Ireland, IEEE,
     7372011.
     738doi:10.1109/INM.2011.5990568.
     739<a href="http://dx.doi.org/10.1109/INM.2011.5990568">http://dx.doi.org/10.1109/INM.2011.5990568</a>
     740<br><br><b>Abstract: </b>Great Plains Environment for Network Innovation (GpENI) is an international testbed for future Internet research. A key component of GpENI is programmable network virtualization (GpENI-VINI). The scope of this paper is to present the framework, implementation and integration experience with network virtualization in GpENI. In particular, this is described through our experience of implementing and integrating the XORP (eXtensible Open Router Platform) routing platform into GpENI-VINI. Preliminary results on measurements and validation are presented.
     741</li>
     742<br>
     743
     744
     745
     746<li>
     747<b>Chin, Tommy and Mountrouidou, Xenia and Li, Xiangyang and Xiong, Kaiqi</b>
     748, &quot;Selective Packet Inspection to Detect DoS Flooding Using Software Defined Networking (SDN).&quot;
     749Distributed Computing Systems Workshops (ICDCSW), 2015 IEEE 35th International Conference on, IEEE,
     7502015.
     751doi:10.1109/icdcsw.2015.27.
     752<a href="http://dx.doi.org/10.1109/icdcsw.2015.27">http://dx.doi.org/10.1109/icdcsw.2015.27</a>
     753<br><br><b>Abstract: </b>Software-defined networking (SDN) and Open Flow have been driving new security applications and services. However, even if some of these studies provide interesting visions of what can be achieved, they stop short of presenting realistic application scenarios and experimental results. In this paper, we discuss a novel attack detection approach that coordinates monitors distributed over a network and controllers centralized on an SDN Open Virtual Switch (OVS), selectively inspecting network packets on demand. With different scale of network views and information availability, these two elements collaboratively detect signature constituents of an attack. Therefore, this approach is able to quickly issue an alert against potential threats followed by careful verification for high accuracy, while balancing the workload on the OVS. We have applied this method for detection and mitigation of TCP SYN flood attacks on Global Environment for Network Innovations (GENI). This realistic experimentation has provided us with insightful findings helpful toward a systematic methodology of SDN-supported attack detection and containment.
     754</li>
     755<br>
     756
     757<li>
     758<b>Chin, Tommy and Mountrouidou, Xenia and Li, Xiangyang and Xiong, Kaiqi</b>
     759, &quot;An SDN-supported collaborative approach for DDoS flooding detection and containment.&quot;
     760Military Communications Conference, MILCOM 2015 - 2015 IEEE, IEEE,
     7612015.
     762doi:10.1109/milcom.2015.7357519.
     763<a href="http://dx.doi.org/10.1109/milcom.2015.7357519">http://dx.doi.org/10.1109/milcom.2015.7357519</a>
     764<br><br><b>Abstract: </b>Software Defined Networking (SDN) has the potential to enable novel security applications that support flexible, on-demand deployment of system elements. It can offer targeted forensic evidence collection and investigation of computer network attacks. Such unique capabilities are instrumental to network intrusion detection that is challenged by large volumes of data and complex network topologies. This paper presents an innovative approach that coordinates distributed network traffic Monitors and attack Correlators supported by Open Virtual Switches (OVS). The Monitors conduct anomaly detection and the Correlators perform deep packet inspection for attack signature recognition. These elements take advantage of complementary views and information availability on both the data and control planes. Moreover, they collaboratively look for network flooding attack signature constituents that possess different characteristics in the level of information abstraction. Therefore, this approach is able to not only quickly raise an alert against potential threats, but also follow it up with careful verification to reduce false alarms. We experiment with this SDN-supported collaborative approach to detect TCP SYN flood attacks on the Global Environment for Network Innovations (GENI), a realistic virtual testbed. The response times and detection accuracy, in the context of a small to medium corporate network, have demonstrated its effectiveness and scalability.
     765</li>
     766<br>
     767
     768
     769
     770<li>
     771<b>Chowdhury and Boutaba, Raouf</b>
     772, &quot;A survey of network virtualization.&quot;
     773Computer Networks,
     7742010.
     775doi:http://dx.doi.org/10.1016/j.comnet.2009.10.017.
     776<a href="http://www.sciencedirect.com/science/article/pii/S1389128609003387">http://www.sciencedirect.com/science/article/pii/S1389128609003387</a>
     777<br><br><b>Abstract: </b>Due to the existence of multiple stakeholders with conflicting goals and policies, alterations to the existing Internet architecture are now limited to simple incremental updates; deployment of any new, radically different technology is next to impossible. To fend off this ossification, network virtualization has been propounded as a diversifying attribute of the future inter-networking paradigm. By introducing a plurality of heterogeneous network architectures cohabiting on a shared physical substrate, network virtualization promotes innovations and diversified applications. In this paper, we survey the existing technologies and a wide array of past and state-of-the-art projects on network virtualization followed by a discussion of major challenges in this area.
     778</li>
     779<br>
     780
     781
     782
     783<li>
     784<b>Collings, Jake and Liu, Jun</b>
     785, &quot;An OpenFlow-Based Prototype of SDN-Oriented Stateful Hardware Firewalls.&quot;
     786Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     7872014.
     788doi:10.1109/icnp.2014.83.
     789<a href="http://dx.doi.org/10.1109/icnp.2014.83">http://dx.doi.org/10.1109/icnp.2014.83</a>
     790<br><br><b>Abstract: </b>This paper describes an Open Flow-based prototype of a SDN-oriented stateful hardware firewall. The prototype of a SDN-oriented stateful hardware firewall includes an Open Flow-enabled switch and a firewall controller. The security rules are specified in the flow table in both the Open Flow-enabled switch and the firewall controller. The firewall controller is in charge of making control decisions on regulating the unidentified traffic flows. A communication channel is needed between a firewall controller and an Open Flow enabled switch. Through this channel, a switch sends to the controller with the information of unidentified flows, and the controller sends to the switch with the control decisions. Constraining this communication overhead is important to the applicability of the prototype because a high communication overhead could disturb the performance evaluation on the operation of a SDN-oriented stateful hardware firewall.
     791</li>
     792<br>
     793
     794
     795
     796<li>
     797<b>Dane, L. and Gurkan, D.</b>
     798, &quot;GENI with a Network Processing Unit: Enriching SDN Application Experiments.&quot;
     799Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     8002014.
     801doi:10.1109/gree.2014.27.
     802<a href="http://dx.doi.org/10.1109/gree.2014.27">http://dx.doi.org/10.1109/gree.2014.27</a>
     803<br><br><b>Abstract: </b>This paper reports the integration of Dell's specialized split data plane (SDP) OpenFlow switch into the GENI testbed. In addition, the paper outlines the research directions in network science and engineering that such a switch may enable together with a new perspective on education in network programming. An SDP switch can be used to perform some specialized processing on flows with special hardware accelerators in addition to hosting any application (running on a Linux OS) that a user may insert on the path of a flow. The SDP switch is composed of a Dell switch (PowerConnect 7024) with an internal physical connection to a sub-unit, Network Processor Unit (NPU), by Cavium Networks. Hosting an OpenvSwitch on the NPU with open hosting of Linux applications enables software-defined networking experiments. The integration challenges/process associated with this unit is presented as a future reference to other such foreign box integrations.
     804</li>
     805<br>
     806
     807
     808
     809<li>
     810<b>Das, S. and Yiakoumis, Y. and Parulkar, G. and McKeown, N. and Singh, P. and Getachew, D. and Desai, P. D.</b>
     811, &quot;Application-aware aggregation and traffic engineering in a converged packet-circuit network.&quot;
     812Optical Fiber Communication Conference and Exposition (OFC/NFOEC), 2011 and the National Fiber Optic Engineers Conference, IEEE,
     8132011.
     814
     815<a href="http://ieeexplore.ieee.org/xpls/abs&#x005F;all.jsp?arnumber=5875210">http://ieeexplore.ieee.org/xpls/abs&#x005F;all.jsp?arnumber=5875210</a>
     816<br><br><b>Abstract: </b>We demonstrate a converged OpenFlow enabled packet-circuit network, where circuit flow properties (guarantee d bandwidth, low latency, low jitter, bandwidth-on-demand, fast recovery) provide differential treatment to dynamically aggregated packet flows for voice, video and web traffic.
     817</li>
     818<br>
     819
     820
     821
     822<li>
     823<b>Deng, Juan and Brooks, Richard R. and Martin, James</b>
     824, &quot;Assessing the Effect of WiMAX System Parameter Settings on MAC-level Local DoS Vulnerability.&quot;
     825International Journal of Performability Engineering,
     8262012.
     827
     828
     829<br><br><b>Abstract: </b>The research community has established that WiMAX networks suffer from Denial of Service (DoS) vulnerabilities. In this paper, we analyze how WiMAX system parameter settings increase or decrease DoS vulnerabilities of WiMAX networks. The behavior of the WiMAX MAC level protocol is sensitive to the settings of core system parameters. Unlike traditional network-based DoS attacks, attacks resulting from parameter misconfiguration are difficult for network operators to detect. We focus on bandwidth contention resolution aspects of the WiMAX MAC protocol. Simulations are performed using the ns-2 simulator. Analysis of Variance (ANOVA) techniques on the resulting simulation data identify which bandwidth contention resolution parameter combinations are crucial for configuring WiMAX to be less vulnerable to DoS attacks.
     830</li>
     831<br>
     832
     833
     834
     835<li>
     836<b>Dong, Mo and Li, Qingxi and Zarchy, Doron and Godfrey, P. Brighten and Schapira, Michael</b>
     837, &quot;PCC: Re-architecting Congestion Control for Consistent High Performance.&quot;
     83812th USENIX Symposium on Networked Systems Design and Implementation (NSDI 15), USENIX Association, Oakland, CA,
     8392015.
     840
     841<a href="https://www.usenix.org/conference/nsdi15/technical-sessions/presentation/dong">https://www.usenix.org/conference/nsdi15/technical-sessions/presentation/dong</a>
     842
     843</li>
     844<br>
     845
     846
     847
     848<li>
     849<b>Duerig, Jonathon and Ricci, Robert and Stoller, Leigh and Strum, Matt and Wong, Gary and Carpenter, Charles and Fei, Zongming and Griffioen, James and Nasir, Hussamuddin and Reed, Jeremy and Wu, Xiongqi</b>
     850, &quot;Getting started with GENI: a user tutorial.&quot;
     851SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
     8522012.
     853doi:10.1145/2096149.2096161.
     854<a href="http://dx.doi.org/10.1145/2096149.2096161">http://dx.doi.org/10.1145/2096149.2096161</a>
     855<br><br><b>Abstract: </b>GENI, the Global Environment for Network Innovations, is a National Science Foundation project to create a &#x76;&#x0308;irtual laboratory at the frontiers of network science and engineering for exploring future internets at scale.&#x20;&#x0308;It provides researchers, educators, and students with resources that they can use to build their own networks that span the country and - through federation - the world. GENI enables experimenters to try out bold new network architectures and designs for networked systems, and to deploy and evaluate these systems on a diverse set of resources over a large footprint. This tutorial is a starting point for running experiments on GENI. It provides an overview of GENI and covers the process of creating a network and running a simple experiment using two tools: the Flack GUI and the INSTOOLS instrumentation service.
     856</li>
     857<br>
     858
     859
     860
     861<li>
     862<b>Duerig, Jonathon and Ricci, Robert and Stoller, Leigh and Wong, Gary and Chikkulapelly, Srikanth and Seok, Woojin</b>
     863, &quot;Designing a Federated Testbed as a Distributed System.&quot;
     8648th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TRIDENTCOM 2012),
     8652012.
     866
     867
     868<br><br><b>Abstract: </b>Traditionally, testbeds for networking and systems research have been stand-alone facilities: each is owned and operated by a single administrative entity, and is intended to be used independently of other testbeds. However, this isolated facility model is at odds with researchers' ever-increasing needs for experiments at larger scale and with a broader diversity of network technologies. The research community will be much better served by a federated model. In this model, each federated testbed maintains its own autonomy and unique strengths, but all federates work together to make their resources available under a common framework. Our challenge, then, is to design a federated testbed framework that balances competing needs: We must establish trust, but at the same time maintain the autonomy of each federated facility. While providing a unified interface to a broad set of resources, we need to expose the diversity that makes them valuable. Finally, our federation should work smoothly in a coordinated fashion, but avoid central points of failure and inter-facility dependencies. We argue that treating testbed design as a federated distributed systems problem is an effective approach to achieving this balance. The technique is illustrated through the example of ProtoGENI, a system we have designed, built, and operated according to the federated model.
     869</li>
     870<br>
     871
     872
     873
     874<li>
     875<b>Dumba, Braulio and Sun, Guobao and Mekky, Hesham and Zhang, Zhi-Li</b>
     876, &quot;Experience in Implementing &#x0026;amp; Deploying a Non-IP Routing Protocol VIRO in GENI.&quot;
     877Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     8782014.
     879doi:10.1109/icnp.2014.85.
     880<a href="http://dx.doi.org/10.1109/icnp.2014.85">http://dx.doi.org/10.1109/icnp.2014.85</a>
     881<br><br><b>Abstract: </b>In this paper, we describe our experience in implementing a non-IP routing protocol - Virtual Id Routing (VIRO) - using the OVS-SDN platform in GENI. As a novel, &#x70;&#x0308;lug-&#x0026;amp;-play&#x2c;&#x0308; routing paradigm for future dynamic networks, VIRO decouples routing/forwarding from addressing by introducing a topology-aware, structured virtual id layer to encode the locations of switches and devices in the physical topology for scalable and resilient routing. Despite its general &#x6d;&#x0308;atch-action&#x20;&#x0308;forwarding function, the existing OVS-SDN platform is closely tied to the conventional Ethernet/IP/TCP header formats, and cannot be directly used to implement the new VIRO routing/forwarding paradigm. As a result, we repurpose the Ethernet MAC address to represent VIRO virtual id, modify and extend the OVS (both within the user space and the kernel space) to implement the VIRO forwarding functions. We also utilize a set of local POX controllers (one per VIRO switch) to emulate the VIRO distributed control plane and one global POX controller to realize the VIRO (centralized) management plane. We evaluate our prototype implementation through the Mininet emulation and GENI deployment test and discuss some lessons learned using the test-bed.
     882</li>
     883<br>
     884
     885
     886
     887<li>
     888<b>Edwards, Sarah and Liu, Xuan and Riga, Niky</b>
     889, &quot;Creating Repeatable Computer Science and Networking Experiments on Shared, Public Testbeds.&quot;
     890SIGOPS Oper. Syst. Rev., ACM, New York, NY, USA,
     8912015.
     892doi:10.1145/2723872.2723884.
     893<a href="http://dx.doi.org/10.1145/2723872.2723884">http://dx.doi.org/10.1145/2723872.2723884</a>
     894<br><br><b>Abstract: </b>There are many compelling reasons to use a shared, public testbed such as GENI, Emulab, or PlanetLab to conduct experiments in computer science and networking. These testbeds support creating experiments with a large and diverse set of resources. Moreover these testbeds are constructed to inherently support the repeatability of experiments as required for scientifically sound research. Finally, the artifacts needed for a researcher to repeat their own experiment can be shared so that others can readily repeat the experiment in the same environment. However using a shared, public testbed is different from conducting experiments on resources either owned by the experimenter or someone the experimenter knows. Experiments on shared, public testbeds are more likely to use large topologies, use scarce resources, and need to be tolerant to outages and maintenances in the testbed. In addition, experimenters may not have access to low-level debugging information. This paper describes a methodology for new experimenters to write and deploy repeatable and sharable experiments which deal with these challenges by: having a clear plan; automating the execution and analysis of an experiment by following best practices from software engineering and system administration; and building scalable experiments. In addition, the paper describes a case study run on the GENI testbed which illustrates the methodology described.
     895</li>
     896<br>
     897
     898
     899
     900<li>
     901<b>El Alaoui, Sara</b>
     902, &quot;Routing Optimization in Interplanetary Networks (Master's Thesis).&quot;
     903
     9042015.
     905
     906<a href="http://scholar.google.com/scholar&#x005F;url?url=http://digitalcommons.unl.edu/cgi/viewcontent.cgi&#x0025;3Farticle&#x0025;3D1110&#x0025;26context&#x0025;3Dcomputerscidiss&#x0026;&#x0023;38;hl=en&#x0026;&#x0023;38;sa=X&#x0026;&#x0023;38;scisig=AAGBfm3bqGZQbbqEX7SG7r5YDIw5epl3sg&#x0026;&#x0023;38;nossl=1&#x0026;&#x0023;38;oi=scholaralrt">http://scholar.google.com/scholar&#x005F;url?url=http://digitalcommons.unl.edu/cgi/viewcontent.cgi&#x0025;3Farticle&#x0025;3D1110&#x0025;26context&#x0025;3Dcomputerscidiss&#x0026;&#x0023;38;hl=en&#x0026;&#x0023;38;sa=X&#x0026;&#x0023;38;scisig=AAGBfm3bqGZQbbqEX7SG7r5YDIw5epl3sg&#x0026;&#x0023;38;nossl=1&#x0026;&#x0023;38;oi=scholaralrt</a>
     907<br><br><b>Abstract: </b>Interplanetary Internet or Interplanetary Networking (IPN) is envisaged as a space network which interconnects spacecrafts, satellites, rovers and orbiters of different planets and comets for efficient exchange of scientific data such as telemetry and images. IPNs are classified among challenged networks because of the unpredictable changes in the network and the large varying delays in communication. These net- works are hard to model using static graphs and do not behave optimally when operated using the static networks' standards and techniques. Delay Tolerant Networking (DTN), in its different implementations, is one of the suggested solutions to overcome these networks' challenges. DTN has different routing techniques, among which Contact Graph Routing (CGR) is the more widely used in IPNs. In this thesis, we identify the shortcoming of CGR that results from overlooking the future contacts, and we propose the Earliest Arrival Optimal Delivery Ratio (EAODR) Routing that examines all the paths both with the desired earliest departure time and in the future in order to choose the earliest arrival path from a given node. EAODR finds the route that delivers the exchanged message (a. k. a. bundle) at most at the same time as CGR's route. In order to do that, we propose a Modified Temporal Graph (MTG) model that provides a near-real-time representation of the deterministic dynamic networks. We base EAODR routing algorithm on the MTG model. Our results show that we can reduce the delay by 12.9&#x0025; compared to CGR when we apply our algorithm to over 50 combinations of bundle sizes and transmission times.
     908</li>
     909<br>
     910
     911
     912
     913<li>
     914<b>El Alaoui, Sara and Palusa, Saichand and Ramamurthy, Byrav</b>
     915, &quot;The Interplanetary Internet Implemented on the GENI Testbed.&quot;
     9162015 IEEE Global Communications Conference (GLOBECOM), IEEE,
     9172015.
     918doi:10.1109/glocom.2014.7417313.
     919<a href="http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&#x0026;&#x0023;38;arnumber=7417313&#x0026;&#x0023;38;isnumber=7416057">http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&#x0026;&#x0023;38;arnumber=7417313&#x0026;&#x0023;38;isnumber=7416057</a>
     920<br><br><b>Abstract: </b>Interplanetary Internet or Interplanetary Networking is envisaged as a space network which interconnects spacecrafts, satellites, rovers and orbiters of different planets and comets for efficient exchange of scientific data such as telemetry and images. In this paper, we implement a layout of the Interplanetary Internet (IPN) with the Interplanetary Overlay Network (ION) software module that uses Contact Graph Routing (CGR). The experiments are then implemented on the Global Environment for Network Innovations (GENI) testbed. Along with realistic contact plans (CP) of the nodes, this network implementation was used to run experiments testing the performance of Delay Tolerant Networking (DTN) with and without cross links between Mars orbiters. The experiments showed that in an Earth-Mars communication network using two Mars orbiters, allowing cross links between the orbiters results in increasing the amount of data transferred by roughly 9.2&#x0025;. Data sent from Mars Rover to the Earth stations also increases by 35.7&#x0025; when a third satellite (Mars Express) was added to the network without cross links. Finally, when cross links are allowed across all satellites orbiting Mars and serving as relay nodes between the Earth stations and Mars rover, the communication was enhanced by almost 46&#x0025;. We conclude that by adding cross links, the performance of the network is enhanced for a better transmission of data from Mars to the Earth, which is very pertinent for the scalability of the network.
     921</li>
     922<br>
     923
     924
     925
     926<li>
     927<b>Elliott, Chip and Falk, Aaron</b>
     928, &quot;An update on the GENI project.&quot;
     929SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
     9302009.
     931doi:10.1145/1568613.1568620.
     932<a href="http://dx.doi.org/10.1145/1568613.1568620">http://dx.doi.org/10.1145/1568613.1568620</a>
     933<br><br><b>Abstract: </b>Environment for Network Innovations. Early prototypes of GENI are starting to come online as an end-to-end system and network researchers are invited to participate by engaging in the design process or using GENI to conduct experiments.
     934</li>
     935<br>
     936
     937
     938
     939<li>
     940<b>Elliott, Steven D.</b>
     941, &quot;Exploring the Challenges and Opportunities of Implementing Software-Defined Networking in a Research Testbed (Master's thesis).&quot;
     942
     9432015.
     944
     945<a href="http://repository.lib.ncsu.edu/ir/bitstream/1840.16/10164/1/etd.pdf">http://repository.lib.ncsu.edu/ir/bitstream/1840.16/10164/1/etd.pdf</a>
     946<br><br><b>Abstract: </b>Designing a new network and upgrading existing network infrastructure are complex and arduous tasks. These projects are further complicated in campus, regional, and international research networks given the large bandwidth and unique segmentation requirements coupled with the unknown implications of testing new network protocols. The software-defined networking (SDN) revolution promises to alleviate these challenges by separating the network control plane from the data plane [208]. This allows for a more flexible and programmable network. While SDN has delivered large dividends to early adopters, it is still a monumental undertaking to re-architect an existing network to use new technology. To ease the transition burden, many research networks have chosen either a hybrid SDN solution or a clean-slate approach. Unfortunately, neither of these approaches can avoid the limitations of existing SDN implementations. For example, software-defined networking can introduce an increase in packet delay in a previously low-latency network. Therefore, it is vital for administrators to have an indepth understanding of these new challenges during the SDN transition. OpenFlow (OF) [209], the protocol many SDN controllers use to communicate with network devices, also has several drawbacks that network architects need to discern before designing the network. Therefore, care must be taken when designing and implementing a software-defined network. This thesis takes an in-depth look at Stanford University, GENI, and OFELIA as case study examples of campus, national, and international research networks that utilize SDN concepts. Additionally, we detail the planning of the future MCNC SDN that will connect several North Carolina research institutions using a high-speed software-defined network. After dissecting the design and implementation of these software-defined research networks, we present common challenges and lessons learned. Our analysis uncovered some common issues in existing software-defined networks. For example, there are problems with the Spanning Tree Protocol (STP), switch/OpenFlow compatibility, hybrid OpenFlow/legacy switch implementations, and the FlowVisor network slicing tool. These potential issues are discussed in detail. Trends include implementation of OpenFlow version 1.3, use of commercial-quality controllers, and a transition to inexpensive network hardware through the use of software switches and NetFPGAs. We hope the findings presented in this thesis will allow network architects to avoid some of the difficulties that arise in design, implementation, and policy decisions when campus and other research networks are transitioning to a software-defined approach.
     947</li>
     948<br>
     949
     950
     951
     952<li>
     953<b>Erazo, Miguel A. and Liu, Jason</b>
     954, &quot;On enabling real-time large-scale network simulation in GENI: the PrimoGENI approach.&quot;
     955Proceedings of the 3rd International ICST Conference on Simulation Tools and Techniques, Torremolinos, Malaga, Spain, ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), ICST, Brussels, Belgium, Belgium,
     9562010.
     957doi:10.4108/ICST.SIMUTOOLS2010.8636.
     958<a href="http://dx.doi.org/10.4108/ICST.SIMUTOOLS2010.8636">http://dx.doi.org/10.4108/ICST.SIMUTOOLS2010.8636</a>
     959<br><br><b>Abstract: </b>The Global Environment for Network Innovations (GENI) is a community-driven research and development effort to build a collaborative and exploratory network experimentation platform, a &#x76;&#x0308;irtual laboratory&#x20;&#x0308;for the design, implementation and evaluation of future Internets. In this paper, we present an overview of PrimoGENI, a GENI project with the goal of extending the GENI suite of interoperable infrastructure to allow network experiments at scale, involving physical, simulated and emulated network entities.
     960</li>
     961<br>
     962
     963
     964
     965<li>
     966<b>Erazo, Miguel A. and Rong, Rong and Liu, Jason</b>
     967, &quot;Symbiotic Network Simulation and Emulation.&quot;
     968ACM Trans. Model. Comput. Simul., ACM, New York, NY, USA,
     9692015.
     970doi:10.1145/2717308.
     971<a href="http://dx.doi.org/10.1145/2717308">http://dx.doi.org/10.1145/2717308</a>
     972<br><br><b>Abstract: </b>A testbed capable of representing detailed operations of complex applications under diverse network conditions is invaluable for understanding the design and performance of new protocols and applications before their real deployment. We introduce a novel method that combines high-performance large-scale network simulation and high-fidelity network emulation, and thus enables real instances of network applications and protocols to run in real operating environments and be tested under simulated network settings. Using our approach, network simulation and emulation can form a symbiotic relationship, through which they are synchronized for an accurate representation of the network-scale traffic behavior. We introduce a model downscaling method along with an efficient queuing model and a traffic reproduction technique, which can significantly reduce the synchronization overhead and improve accuracy. We validate our approach with extensive experiments via simulation and with a real-system implementation. We also present a case study using our approach to evaluate a multipath data transport protocol.
     973</li>
     974<br>
     975
     976
     977
     978<li>
     979<b>Esposito, Flavio and Wang, Yuefeng and Matta, Ibrahim and Day, John</b>
     980, &quot;Dynamic Layer Instantiation as a Service.&quot;
     981Lombard, IL, USENIX Association, Berkeley, CA, USA,
     9822013.
     983
     984<a href="https://www.usenix.org/system/files/nsdip13-paper11.pdf">https://www.usenix.org/system/files/nsdip13-paper11.pdf</a>
     985<br><br><b>Abstract: </b>We demonstrate the dynamic layer instantiation feature of RINA by creating on the fly a new, higher level Virtual Private Cloud DIF. The demonstration includes two IPC processes, VPC1 and VPC2, that initially use two separate private DIFs — an Enterprise DIF and a Cloud Provider DIF — to communicate with their respective local processes. Later on, an enterprise application process App1 asks for a flow service, so as to communicate with App2, a remote application process on the Cloud Provider DIF. Such request, handled by the underlying communication process VPC1, cannot occur unless there is a common underlying Virtual Private Cloud DIF to which both VPC1 and VPC2 subscribe. We demonstrate this dynamic instantiation of the DIF layer over the GENI testbed.
     986</li>
     987<br>
     988
     989
     990
     991<li>
     992<b>Feamster, Nick and Gao, Lixin and Rexford, Jennifer</b>
     993, &quot;How to lease the internet in your spare time.&quot;
     994SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
     9952007.
     996doi:10.1145/1198255.1198265.
     997<a href="http://doi.acm.org/10.1145/1198255.1198265">http://doi.acm.org/10.1145/1198255.1198265</a>
     998<br><br><b>Abstract: </b>Today's Internet Service Providers (ISPs) serve two roles: managing their network infrastructure and providing (arguably limited) services to end users. We argue that coupling these roles impedes the deployment of new protocols and architectures, and that the future Internet should support two separate entities: infrastructure providers (who manage the physical infrastructure) and service providers (who deploy network protocols and offer end-to-end services). We present a high-level design for Cabo, an architecture that enables this separation; we also describe challenges associated with realizing this architecture
     999</li>
     1000<br>
     1001
     1002
     1003
     1004<li>
     1005<b>Feamster, Nick and Nayak, Ankur and Kim, Hyojoon and Clark, Russell and Mundada, Yogesh and Ramachandran, Anirudh and bin Tariq, Mukarram</b>
     1006, &quot;Decoupling policy from configuration in campus and enterprise networks.&quot;
     10072010 17th IEEE Workshop on Local &#x0026; Metropolitan Area Networks (LANMAN), Long Branch, NJ, USA, IEEE,
     10082010.
     1009doi:10.1109/LANMAN.2010.5507162.
     1010<a href="http://dx.doi.org/10.1109/LANMAN.2010.5507162">http://dx.doi.org/10.1109/LANMAN.2010.5507162</a>
     1011<br><br><b>Abstract: </b>This paper surveys our ongoing work on the use of software-defined networking to simplify two acute policy problems in campus and enterprise network operations: access control and information flow control. We describe how the current coupling of high-level policy with low-level configuration makes these problems challenging today. We describe the specific policy problems faced by campus and enterprise network operators; illustrate our approach, which leverages recent trends in separating the network's ” control plane” from the data plane; and show how this approach can be applied to simplify these two enterprise network management tasks. We also describe our ongoing deployment efforts to build a campus network testbed where trial designs can be deployed and evaluated. We close with a summary of current and future research challenges for solving challenges within enterprise networks within the context of this new paradigm.
     1012</li>
     1013<br>
     1014
     1015
     1016
     1017<li>
     1018<b>Fei, Zongming and Xu, Qingrong and Lu, Hui</b>
     1019, &quot;Generating large network topologies for GENI experiments.&quot;
     1020SOUTHEASTCON 2014, IEEE, IEEE,
     10212014.
     1022doi:10.1109/secon.2014.6950726.
     1023<a href="http://dx.doi.org/10.1109/secon.2014.6950726">http://dx.doi.org/10.1109/secon.2014.6950726</a>
     1024<br><br><b>Abstract: </b>The Global Environment for Network Innovations (GENI) is a virtual laboratory which provides the infrastructure and resources for setting up network experiments. At present, GENI experimenters need to draw the topology in detail with a tool such as Flack, describing every node and every link in the experiment. This is not a problem for small-scale experiments. However, if an experiment needs a large-scale network topology, it is difficult for experimenters to accomplish the task. To deal with the problem, this paper develops a web application that can create large-scale network topologies in the GENI environment automatically. It makes use of existing network topology generators, such as GT-ITMand INET, and adapts them to be used in the GENI environment. The system can interface with GENI seamlessly. With the tool, the task of setting up large-scale experiments by GENI experimenters is made as easy as simply specifying high-level parameters of the topology.
     1025</li>
     1026<br>
     1027
     1028
     1029
     1030<li>
     1031<b>Fei, Zongming and Yi, Ping and Yang, Jianjun</b>
     1032, &quot;A Performance Perspective on Choosing between Single Aggregate and Multiple Aggregates for GENI Experime nts.&quot;
     1033EAI Endorsed Transactions on Industrial Networks and Intelligent Systems,
     10342014.
     1035doi:10.4108/inis.1.1.e5.
     1036<a href="http://dx.doi.org/10.4108/inis.1.1.e5">http://dx.doi.org/10.4108/inis.1.1.e5</a>
     1037<br><br><b>Abstract: </b>The Global Environment for Network Innovations (GENI) provides a virtual laboratory for exploring future internets at scale. It consists of many geographically distributed aggregates for providing computing and networking resources for setting up network experiments. A key design question for GENI experimenters is where they should reserve the resources, and in particular whether they should reserve the resources from a single aggregate or from multiple aggregates. This not only depends on the nature of the experiment, but needs a better understanding of underlying GENI networks as well. This paper studies the performance of GENI networks, with a focus on the tradeoff between single aggregate and multiple aggregates in the design of GENI experiments from the performance perspective. The analysis of data collected will shed light on the decision process for designing GENI experiments.
     1038</li>
     1039<br>
     1040
     1041
     1042
     1043<li>
     1044<b>Femminella, Mauro and Francescangeli, Roberto and Reali, Gianluca and Lee, Jae W. and Schulzrinne, Henning</b>
     1045, &quot;An enabling platform for autonomic management of the future internet.&quot;
     1046IEEE Network,
     10472011.
     1048doi:10.1109/MNET.2011.6085639.
     1049<a href="http://dx.doi.org/10.1109/MNET.2011.6085639">http://dx.doi.org/10.1109/MNET.2011.6085639</a>
     1050<br><br><b>Abstract: </b>This article shows an autonomic management solution based on the recently defined programmable node architecture NetServ. The article starts with a general description of the classical network management requirements and their adaptation to the expected network evolution. After a description of the major issues characterizing the management of the expected Future Internet, the main autonomic management paradigms, and some recently introduced autonomic service platforms, we show and demonstrate the effectiveness of the NetServ architecture. Born as a means to deploy and execute networked services at runtime over programmable routers, NetServ has proved to be a suitable environment for hosting an autonomic management architecture.
     1051</li>
     1052<br>
     1053
     1054
     1055
     1056<li>
     1057<b>Fund, Fraida and Dong, Chen and Korakis, Thanasis and Panwar, Shivendra</b>
     1058, &quot;A Framework for Multidimensional Measurements on an Experimental WiMAX Testbed.&quot;
     1059Testbeds and Research Infrastructure. Development of Networks and Communities, Springer Berlin Heidelberg,
     10602012.
     1061doi:10.1007/978-3-642-35576-9&#x005F;32.
     1062<a href="http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;32">http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;32</a>
     1063<br><br><b>Abstract: </b>A major difficulty in the design, study, and implementation of wireless protocols and applications is the multitude of nondeterministic factors (e.g. interference, weather conditions, competing traffic) that can affect their performance. For this reason, testbeds that enable researchers to quantify these influences have become increasingly essential in the wireless research community. The growing sophistication of wireless testbeds and the wide array of services they can provide to researchers have advanced the field tremendously. Toward this end, we present an early implementation of an instrumentation and measurement framework that we have deployed on an open-access 802.16e wireless research testbed at the Polytechnic Institute of NYU. We have created a set of tools to allow experimenters to routinely collect measurements of environmental conditions during experiment runtime. These tools integrate high volumes of multidimensional measurement data from a diverse array of sources, including measurements from software defined radio peripherals, sensors, and network device drivers. With this, we aim to give researchers the ability to conduct rigorous and repeatable over-the-air experiments. We also foresee potential applications for this framework beyond its use in experiments, such as in long-term testbed monitoring.
     1064</li>
     1065<br>
     1066
     1067
     1068
     1069<li>
     1070<b>Fund, Fraida and Wang, Cong and Korakis, Thanasis and Zink, Michael and Panwar, Shivendra</b>
     1071, &quot;GENI WiMAX Performance: Evaluation and Comparison of Two Campus Testbeds.&quot;
     10722013 Proceedings Second GENI Research and Educational Experiment Workshop, Salt Lake City, UT, IEEE,
     10732013.
     1074doi:10.1109/GREE.2013.23.
     1075<a href="http://dx.doi.org/10.1109/GREE.2013.23">http://dx.doi.org/10.1109/GREE.2013.23</a>
     1076<br><br><b>Abstract: </b>In the last few years, there has been an increasing awareness of the need to evaluate new mobile applications and protocols in realistic wireless settings, and platforms such as the GENI WiMAX testbeds have been developed to fulfill this need. However, wireless testbed users have experienced frustration when straightforward usage scenarios do not consistently agree with the high data rates that are advertised by the wireless technology. This work seeks to clarify the performance characteristics of two GENI WiMAX testbeds under various wireless signal conditions and network traffic patterns. By measuring the performance of several popular wireless Internet applications in two very different wireless environments, we gain a deeper understanding of how a researcher may expect the GENI WiMAX platform to behave. Our findings include some counterintuitive results, e.g. that increasing signal quality can reduce application throughput, and that applications using a single TCP flow may achieve as much as 72&#x0025; less throughput than an application in an identical setting that uses multiple TCP flows. With this work, we hope to help other researchers design realistic experiments on wireless Internet systems, understand the perceived shortcomings of the GENI WiMAX platform, and interpret their experimental results in the context of the wireless setting in which the experiment was conducted.
     1077</li>
     1078<br>
     1079
     1080
     1081
     1082<li>
     1083<b>Gangam, Sriharsha and Blanton, Ethan and Fahmy, Sonia</b>
     1084, &quot;Exercises for Graduate Students using GENI.&quot;
     1085First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     10862012.
     1087
     1088
     1089<br><br><b>Abstract: </b>GENI brings together a wide variety of heterogeneous networking infrastructure and technologies under a common platform. We propose programming exercises for graduate students to introduce GENI and enable students to conduct high fidelity networking experiments. In this paper, we focus on an exercise to study congestion control and reliability using the ProtoGENI aggregate. A planned second exercise aims to leverage GENI OpenFlow aggregates to study firewalls and QoS mechanisms. We believe that these lab exercises will expose students to key networking concepts and recent research directions, e.g., in the data center context.
     1090</li>
     1091<br>
     1092
     1093
     1094
     1095<li>
     1096<b>Gangam, Sriharsha and Fahmy, Sonia</b>
     1097, &quot;Mitigating interference in a network measurement service.&quot;
     10982011 IEEE Nineteenth IEEE International Workshop on Quality of Service, San Jose, CA, USA, IEEE,
     10992011.
     1100doi:10.1109/IWQOS.2011.5931347.
     1101<a href="http://dx.doi.org/10.1109/IWQOS.2011.5931347">http://dx.doi.org/10.1109/IWQOS.2011.5931347</a>
     1102<br><br><b>Abstract: </b>Shared measurement services offer key advantages over conventional ad-hoc techniques for network monitoring. A measurement service may receive measurement requests concurrently from different applications and network administrators. These measurement requests are often served by injecting active network measurement traffic between two hosts. Two active measurements are said to interfere when the probe packets of one measurement tool are viewed as network traffic by the other. This may lead to faulty measurement readings. In this paper, we model the measurement interference problem, and show how to schedule measurement tasks to reduce interference and hence increase measurement accuracy. We propose twelve computationally tractable algorithms that decrease the total completion time (makespan) of measurement tasks, while avoiding interference. Our evaluation shows that the algorithm we refer to as Largest Area First, Busiest Node First - Earliest Interval Schedule (LAFBNF-EIS) has a mean makespan of about 5&#x0025; more than the theoretical lower bound over our set of measurement workloads.
     1103</li>
     1104<br>
     1105
     1106
     1107
     1108<li>
     1109<b>Gao, Jingcheng and Xiao, Yang</b>
     1110, &quot;ProtoGENI DoS/DDoS Security Tests and Experiments.&quot;
     1111First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     11122012.
     1113
     1114
     1115<br><br><b>Abstract: </b>his paper will explain some tests and experiments to investigate selected security issues through ProtoGENI mainly during Spiral 3 time period and the beginning of Spiral 4. In this paper, we conduct multiple sets of DoS/ DDoS attacks in the current ProtoGENI testbed. These attacks show that it is very possible that ProtoGENI nodes may render vulnerabilities to such attacks.
     1116</li>
     1117<br>
     1118
     1119
     1120
     1121<li>
     1122<b>Gember, Aaron and Dragga, Chris and Akella, Aditya</b>
     1123, &quot;ECOS: Practical Mobile Application Offloading for Enterprises.&quot;
     11242nd USENIX Workshop on Hot Topics in Management of Internet, Cloud, and Enterprise Networks and Services (Hot-ICE '12),
     11252012.
     1126
     1127<a href="http://www.usenix.org/conference/hot-ice12/ecos-practical-mobile-application-of&#x0025;EF&#x0025;AC&#x0025;82oading-enterprises">http://www.usenix.org/conference/hot-ice12/ecos-practical-mobile-application-of&#x0025;EF&#x0025;AC&#x0025;82oading-enterprises</a>
     1128<br><br><b>Abstract: </b>Offloading has emerged as a promising idea to allow handheld devices to access intensive applications without performance or energy costs. This could be particularly useful for enterprises seeking to run line-of-business applications on handhelds. However, we must address two practical roadblocks in order to make offloading amenable for enterprises: (i) ensuring data privacy and the use of trusted offloading resources, and (ii) accommodating offload at scale with diverse handheld objectives and compute resource capabilities. We present the design and implementation of an Enterprise-Centric Offloading System (ECOS) which augments prior offloading proposals to address these issues. ECOS uses a logically central controller to opportunistically leverage diverse compute resources, while tightly controlling where specific applications offload depending on privacy, performance, and energy constraints of users and applications. A wide range of experiments using a real prototype establish the effectiveness of our approach.
     1129</li>
     1130<br>
     1131
     1132
     1133
     1134<li>
     1135<b>Ghaffarinejad, A. and Syrotiuk, V. R.</b>
     1136, &quot;Load Balancing in a Campus Network Using Software Defined Networking.&quot;
     1137Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     11382014.
     1139doi:10.1109/gree.2014.9.
     1140<a href="http://dx.doi.org/10.1109/gree.2014.9">http://dx.doi.org/10.1109/gree.2014.9</a>
     1141<br><br><b>Abstract: </b>Today, commercial load balancers are often in use, including in the production network at Arizona State University (ASU). One of the main issues such load balancers face is that they use a static scheme for load distribution. However, at particular times of the academic year, such as during course registration, the network exhibits significant variations in both temporal and spatial traffic characteristics. At these times, students experience much greater latency and become frustrated with the network service. To address this problem, our aim is to develop an SDN-based approach to load balancing to better cope with the traffic variation.
     1142</li>
     1143<br>
     1144
     1145
     1146
     1147<li>
     1148<b>Grandl, Robert and Han, Dongsu and Lee, Suk B. and Lim, Hyeontaek and Machado, Michel and Mukerjee, Matthew and Naylor, David</b>
     1149, &quot;Supporting network evolution and incremental deployment with XIA.&quot;
     1150Proceedings of the ACM SIGCOMM 2012 conference on Applications, technologies, architectures, and protocols for computer communication, Helsinki, Finland, ACM, New York, NY, USA,
     11512012.
     1152doi:10.1145/2342356.2342410.
     1153<a href="http://dx.doi.org/10.1145/2342356.2342410">http://dx.doi.org/10.1145/2342356.2342410</a>
     1154<br><br><b>Abstract: </b>eXpressive Internet Architecture (XIA) [1] is an architecture that natively supports multiple communication types and allows networks to evolve their abstractions and functionality to accommodate new styles of communication over time. XIA embeds an elegant mechanism for handling unforeseen communication types for legacy routers. In this demonstration, we show that XIA overcomes three key barriers in network evolution (outlined below) by (1) allowing end-hosts and applications to start using new communication types (e.g., service and content) before the network supports them, (2) ensuring that upgrading a subset of routers to support new functionalities immediately benefits applications, and (3) using the same mechanisms we employ for 1 and 2 to incrementally deploy XIA in IP networks.
     1155</li>
     1156<br>
     1157
     1158
     1159
     1160<li>
     1161<b>Griffioen, J. and Fei, Zongming and Nasir, H. and Wu, Xiongqi and Reed, J. and Carpenter, C.</b>
     1162, &quot;GENI-Enabled Programming Experiments for Networking Classes.&quot;
     1163Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     11642013.
     1165doi:10.1109/gree.2013.30.
     1166<a href="http://dx.doi.org/10.1109/gree.2013.30">http://dx.doi.org/10.1109/gree.2013.30</a>
     1167<br><br><b>Abstract: </b>Although GENI has been readily embraced by the research community as a testbed for exploring new network architectures and services, its use as an educational tool has not seen the same level of acceptance and usage. There are multiple reasons for this, not the least of which is a lack of good examples showing how to use GENI in an educational setting. This paper attempts to remedy this by describing our experiences using GENI in our networking classes at the University of Kentucky. Using GENI as the experimental basis for the projects in our classes allowed us to leverage several of its rich set of features including its global span of resources, programmability, virtualization, and instrumentation and measurement tools. In particular, we describe two projects that we have used in our networking classes, and we share some of the experience we gained in the process. As a result, these experiences motivated us to develop and integrate new functions into the GENI desktop in order to make it easier to access and control GENI's various resources and tools.
     1168</li>
     1169<br>
     1170
     1171<li>
     1172<b>Griffioen, J. and Fei, Zongming and Nasir, H. and Wu, Xiongqi and Reed, J. and Carpenter, C.</b>
     1173, &quot;The design of an instrumentation system for federated and virtualized network testbeds.&quot;
     1174Network Operations and Management Symposium (NOMS), 2012 IEEE, IEEE,
     11752012.
     1176doi:10.1109/NOMS.2012.6212061.
     1177<a href="http://dx.doi.org/10.1109/NOMS.2012.6212061">http://dx.doi.org/10.1109/NOMS.2012.6212061</a>
     1178<br><br><b>Abstract: </b>Much of the GENI effort in developing network testbeds has been focused on building the control frameworks needed to allocate and initialize the network resources that make up an experiment. We argue that building the instrumentation and measurement system to monitor and capture the behavior of the network is just as important and challenging as setting up the network itself, especially in a virtualized and federated environment where getting information from experimental nodes is too complicated and too much to handle for a typical user. In this paper, we describe the design of an instrumentation and measurement infrastructure that allows users to monitor their experiments. The challenge that virtualization and federation of GENI testbeds bring to instrumentation and monitoring is how to hide the details of instrumentation setup from users so that users do not need to be experts in system administration or network management of virtualized and federated systems, but are still able to ” see” what is going on with their experiments. Our instrumentation tool sets up experiment-specific monitoring infrastructure that is tailored to capture, record, and display only information associated with that experiment. Our tools are currently available in GENI, and we present a simple example of how to use them to instrument an experiment.
     1179</li>
     1180<br>
     1181
     1182
     1183
     1184<li>
     1185<b>Griffioen, James and Fei, Zongming and Nasir, Hussamuddin and Wu, Xiongqi and Reed, Jeremy and Carpenter, Charles</b>
     1186, &quot;Measuring experiments in GENI.&quot;
     1187Computer Networks,
     11882014.
     1189doi:10.1016/j.bjp.2013.10.016.
     1190<a href="http://dx.doi.org/10.1016/j.bjp.2013.10.016">http://dx.doi.org/10.1016/j.bjp.2013.10.016</a>
     1191<br><br><b>Abstract: </b>Experimentation with new network architectures and protocols is one of the primary motivations for building future Internet testbeds such as the Global Environment for Network Innovations (GENI) testbed. A key part of experimentation is the ability to observe, measure, evaluate, and compare these new architectures and protocols. Observing an experiment's network performance requires setting up the measurement infrastructure needed to monitor and record the behavior of the network. It also requires a full set of tools and user interfaces that enable access to the measurement data both while the experiment is running and later during post-analysis. To simplify the task of measuring experiments in future Internet testbeds like GENI, we developed an instrumentation and measurement system called INSTOOLS. It automates the process of setting up the measurement infrastructure, tailoring the measurement infrastructure and the data capture to the experimental network's topology and configuration. In addition, INSTOOLS provides a suite of tools via its ” portal” service that make it easy for users to observe, measure, format, and archive data from their experiments. This paper describes the INSTOOLS system and the set of interfaces/tools it offers to users. INSTOOLS has been in use for several years, and we provide performance results that illustrate its scalability. We also present our second-generation portal, the GENI One Stop Portal, that offers a comprehensive interface to a wide range of tools.
     1192</li>
     1193<br>
     1194
     1195
     1196
     1197<li>
     1198<b>Griffioen, James and Fei, Zongming and Nasir, Hussanmuddin and Wu, Xiongqi and Reed, Jeremy and Carpenter, Charles</b>
     1199, &quot;Teaching with the Emerging GENI Network.&quot;
     1200Proceedings of the 2012 International Conference on Frontiers in Education: Computer Science and Computer Engineering (FECS), Las Vegas,
     12012012.
     1202
     1203<a href="http://worldcomp-proceedings.com/proc/p2012/FEC3780.pdf">http://worldcomp-proceedings.com/proc/p2012/FEC3780.pdf</a>
     1204<br><br><b>Abstract: </b>Over the last few years the National Science Foundation (NSF) has been investing in and developing a new network called GENI, a wide-area testbed network for at-scale experimentation with future internet designs. The GENI network has recently become available for use and is beginning to attract users. In this paper, we take a closer look at GENI with a particular focus on how GENI can be used to enhance education in the areas of computer science and computer engineering. We describe what GENI is, the resources available in GENI, and how instructors might use GENI in their classes. Being early adopters, we describe our experience using GENI in our classes, and we point out various features and challenges of using GENI. Finally, we provide tips and pointers to instructors who are interested in incorporating GENI into their own classes.
     1205</li>
     1206<br>
     1207
    681208
    691209
     
    821222
    831223<li>
     1224<b>Guan, Xinjie and Choi, Baek-Young and Song, Sejun</b>
     1225, &quot;Reliability and Scalability Issues in Software Defined Network Frameworks.&quot;
     1226Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     12272013.
     1228doi:10.1109/gree.2013.28.
     1229<a href="http://dx.doi.org/10.1109/gree.2013.28">http://dx.doi.org/10.1109/gree.2013.28</a>
     1230<br><br><b>Abstract: </b>Software Defined Network (SDN) structure has been proposed for its flexibility in deployment and management. As an implementation of SDN structure, OpenFlow protocol decouples data plane and control plane so that flexible and programmable installation and management of forwarding rules are allowed. However, on the other hand, the decoupled structure raises additional computational and network resources consumption that even may lead to fatal disasters. In this study, we examine the issues of reliability and scalability of SDN under disaster scenarios on a GENI test-bed. Observations from our experiments show that more attention should be paid to improve the reliability and scalability of SDN and its frameworks.
     1231</li>
     1232<br>
     1233
     1234
     1235
     1236<li>
     1237<b>Gupta, Arpit and Vanbever, Laurent and Shahbaz, Muhammad and Donovan, Sean P. and Schlinker, Brandon and Feamster, Nick and Rexford, Jennifer and Shenker, Scott and Clark, Russ and Katz-Bassett, Ethan</b>
     1238, &quot;SDX: A Software Defined Internet Exchange.&quot;
     1239Proceedings of the 2014 ACM Conference on SIGCOMM, Chicago, Illinois, USA, ACM, New York, NY, USA,
     12402014.
     1241doi:10.1145/2619239.2626300.
     1242<a href="http://dx.doi.org/10.1145/2619239.2626300">http://dx.doi.org/10.1145/2619239.2626300</a>
     1243<br><br><b>Abstract: </b>BGP severely constrains how networks can deliver traffic over the Internet. Today's networks can only forward traffic based on the destination IP prefix, by selecting among routes offered by their immediate neighbors. We believe Software Defined Networking (SDN) could revolutionize wide-area traffic delivery, by offering direct control over packet-processing rules that match on multiple header fields and perform a variety of actions. Internet exchange points (IXPs) are a compelling place to start, given their central role in interconnecting many networks and their growing importance in bringing popular content closer to end users. To realize a Software Defined IXP (an &#x53;&#x0308;DX&#x29;&#x0308;, we must create compelling applications, such as &#x61;&#x0308;pplication-specific peering&#x2d;&#x0308;--where two networks peer only for (say) streaming video traffic. We also need new programming abstractions that allow participating networks to create and run these applications and a runtime that both behaves correctly when interacting with BGP and ensures that applications do not interfere with each other. Finally, we must ensure that the system scales, both in rule-table size and computational overhead. In this paper, we tackle these challenges and demonstrate the flexibility and scalability of our solutions through controlled and in-the-wild experiments. Our experiments demonstrate that our SDX implementation can implement representative policies for hundreds of participants who advertise full routing tables while achieving sub-second convergence in response to configuration changes and routing updates.
     1244</li>
     1245<br>
     1246
     1247
     1248
     1249<li>
     1250<b>Herron, Jon-Paul</b>
     1251, &quot;GENI Meta-Operations Center.&quot;
     12522008 IEEE Fourth International Conference on eScience, Indianapolis, IN, USA, IEEE,
     12532008.
     1254doi:10.1109/eScience.2008.103.
     1255<a href="http://dx.doi.org/10.1109/eScience.2008.103">http://dx.doi.org/10.1109/eScience.2008.103</a>
     1256<br><br><b>Abstract: </b>NSF's GENI program represents an opportunity to build the kind of programmable, virtualized testbed scientists exploring the future of networking will need to support their research. As with any other scientific instrument, it will be crucial that the GENI infrastructure offer repeatable, consistent results to the researchers using it.The GENI Meta-Operations Center, operated by the Global Research NOC at Indiana University, will develop the software, protocols, and processes needed to ensure the repeatability, consistency, and efficiency of GENI.
     1257</li>
     1258<br>
     1259
     1260
     1261
     1262<li>
     1263<b>Huang, Shu and Xu, Hao and Xin, Yufeng and Brieger, L. and Moore, R. and Rajasekar, A.</b>
     1264, &quot;A Framework for Integration of Rule-Oriented Data Management Policies with Network Policies.&quot;
     1265Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     12662014.
     1267doi:10.1109/gree.2014.19.
     1268<a href="http://dx.doi.org/10.1109/gree.2014.19">http://dx.doi.org/10.1109/gree.2014.19</a>
     1269<br><br><b>Abstract: </b>Traditionally data management software running on top of the Internet has very limited primitives to interact with the networking layer. This limitation has become a major road-block to develop next generation data management applications requiring high-bandwidth and dynamic network configuration. In this work, we present a policy-driven software framework that acts as an adaptation layer between the data management software and SDN networks. This framework allows a tight coupling between the data grid and the network and therefore makes complex workflow-like cross-layer computation possible. We have prototyped this adaptation layer integrated with iRODS, a popular policy-driven data grid software and Floodlight, a popular OpenFlow controller, and demonstrate how network policies become part of the overall data grid policies to improve the application performance.
     1270</li>
     1271<br>
     1272
     1273
     1274
     1275<li>
     1276<b>Huang, Shufeng and Griffioen, J. and Calvert, K. L.</b>
     1277, &quot;Fast-Tracking GENI Experiments Using HyperNets.&quot;
     1278Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     12792013.
     1280doi:10.1109/gree.2013.10.
     1281<a href="http://dx.doi.org/10.1109/gree.2013.10">http://dx.doi.org/10.1109/gree.2013.10</a>
     1282<br><br><b>Abstract: </b>Although the underlying network resources needed to support virtualized networks are rapidly becoming available, the tools and abstractions needed to effectively make use of these virtual networks is severely lacking. Although networks like GENI are now available to experimenters, creating an experimental network can still be a daunting and error-prone task. While virtual networks enable experimenters to build tailored networks from the &#x67;&#x0308;round up&#x2c;&#x0308; starting from scratch is rarely what an experimenter wants to do. Moreover, the challenges of incorporating real-world users into GENI experiments make it difficult to benefit real users or obtain realistic traffic. In this paper we describe a new service designed to simplify the process of setting up and running GENI experiments while at the same time adding support for real-world users to join GENI experiments. Our approach is based on a network hypervisor service used to deploy &#x48;&#x0308;yperNets&#x3a;&#x0308; pre-defined experimental environments that can be quickly and easily created by experimenters. To illustrate the utility and simplicity of our approach, we describe two example HyperNets, and show how our network hypervisor service is able to automatically deploy them on GENI. We then present some initial performance results from our implentation on GENI. Because our network hypervisor is itself a client of GENI (i.e., it calls the GENI AM APIs to create HyperNets), we briefly discuss our experience using GENI and the challenges we encountered mapping HyperNets onto the GENI framework.
     1283</li>
     1284<br>
     1285
     1286
     1287
     1288<li>
     1289<b>Huang, Shufeng and Griffioen, James</b>
     1290, &quot;Network Hypervisors: Managing the Emerging SDN Chaos.&quot;
     1291Computer Communications and Networks (ICCCN), 2013 22nd International Conference on, IEEE,
     12922013.
     1293doi:10.1109/icccn.2013.6614160.
     1294<a href="http://dx.doi.org/10.1109/icccn.2013.6614160">http://dx.doi.org/10.1109/icccn.2013.6614160</a>
     1295<br><br><b>Abstract: </b>Software-Defined Networking (SDN) has been widely recognized as a promising way to deploy new services and protocols in future networks. The programmability and control offered by SDN networks enables users and applications to define virtually every aspect of the network architecture. Unfortunately, this flexibility comes at a cost - a cost that has the potential to significantly limit its adoption. First, in order to offer complete flexibility, today's SDN networks provide low-level API's on which almost any type of service can be written. In the process, it can actually become more difficult to implement the higher level complex services needed by future networks. Second, emerging SDN networks exhibit a heterogeneity reminiscent of the early Internet, with limited ability to piece together the various SDN platforms being deployed. In this paper we propose a new way to construct SDN networks consisting of multiple SDN providers offering virtualizable networking resources across the Internet. At the heart of our approach is a Network Hypervisor service that is capable of internetworking various SDN providers together. Moreover, our Network Hypervisor builds on the low-level APIs provided by SDNs to create a unified set of high-level abstractions and APIs that greatly simplify the task of building and deploying complex network services over SDN.
     1296</li>
     1297<br>
     1298
     1299
     1300
     1301<li>
     1302<b>Huang, Shufeng and Griffioen, James and Calvert, Ken</b>
     1303, &quot;PVNs: Making Virtualized Network Infrastructure Usable.&quot;
     1304ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS '12),
     13052012.
     1306doi:10.1145/2396556.2396590.
     1307<a href="http://dx.doi.org/10.1145/2396556.2396590">http://dx.doi.org/10.1145/2396556.2396590</a>
     1308<br><br><b>Abstract: </b>Network virtualization is becoming a fundamental building block of future Internet architectures. Although the underlying network infrastructure needed to dynamically create and deploy custom virtual networks is rapidly taking shape ( e.g., GENI), constructing and using a virtual network is still a challenging and labor intensive task, one best left to experts. In this paper, we present the concept of a Packaged Virtual Network (PVN), that enables normal users to easily download, deploy and use application-specific virtual networks. At the heart of our approach is a PVN Hypervisor that ” runs” a PVN by allocating the virtual network resources needed by the PVN and then connecting the PVN's participants into the network on demand. To demonstrate our PVN approach, we implemented a multicast PVN that runs on the PVN hypervisor prototype using ProtoGENI as the underlying virtual network, allowing average users to create their own private multicast network.
     1309</li>
     1310<br>
     1311
     1312
     1313
     1314<li>
     1315<b>Javed, Umar and Cunha, Italo and Choffnes, David and Katz-Bassett, Ethan and Anderson, Thomas and Krishnamurthy, Arvind</b>
     1316, &quot;PoiRoot: Investigating the Root Cause of Interdomain Path Changes.&quot;
     1317Proceedings of the ACM SIGCOMM 2013 conference, ACM, New York, NY, USA,
     13182013.
     1319doi:10.1145/2486001.2486036.
     1320<a href="http://dx.doi.org/10.1145/2486001.2486036">http://dx.doi.org/10.1145/2486001.2486036</a>
     1321<br><br><b>Abstract: </b>Interdomain path changes occur frequently. Because routing protocols expose insufficient information to reason about all changes, the general problem of identifying the root cause remains unsolved. In this work, we design and evaluate PoiRoot, a real-time system that allows a provider to accurately isolate the root cause (the network responsible) of path changes affecting its prefixes. First, we develop a new model describing path changes and use it to provably identify the set of all potentially responsible networks. Next, we develop a recursive algorithm that accurately isolates the root cause of any path change. We observe that the algorithm requires monitoring paths that are generally not visible using standard measurement tools. To address this limitation, we combine existing measurement tools in new ways to acquire path information required for isolating the root cause of a path change. We evaluate PoiRoot on path changes obtained through controlled Internet experiments, simulations, and &#x69;&#x0308;n-the-wild&#x20;&#x0308;measurements. We demonstrate that PoiRoot is highly accurate, works well even with partial information, and generally narrows down the root cause to a single network or two neighboring ones. On controlled experiments PoiRoot is 100&#x0025; accurate, as opposed to prior work which is accurate only 61.7&#x0025; of the time.
     1322</li>
     1323<br>
     1324
     1325
     1326
     1327<li>
     1328<b>Jin, Ruofan and Wang, Bing</b>
     1329, &quot;Malware Detection for Mobile Devices Using Software-Defined Networking.&quot;
     1330Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     13312013.
     1332doi:10.1109/gree.2013.24.
     1333<a href="http://dx.doi.org/10.1109/gree.2013.24">http://dx.doi.org/10.1109/gree.2013.24</a>
     1334<br><br><b>Abstract: </b>The rapid adoption of mobile devices comes with the growing prevalence of mobile malware. Mobile malware poses serious threats to personal information and creates challenges in securing network. Traditional network services provide connectivity but do not have any direct mechanism for security protection. The emergence of Software-Defined Networking (SDN) provides a unique opportunity to achieve network security in a more efficient and flexible manner. In this paper, we analyze the behaviors of mobile malware, propose several mobile malware detection algorithms, and design and implement a malware detection system using SDN. Our system detects mobile malware by identifying suspicious network activities through real-time traffic analysis, which only requires connection establishment packets. Specifically, our detection algorithms are implemented as modules inside the OpenFlow controller, and the security rules can be imposed in real time. We have tested our system prototype using both a local testbed and GENI infrastructure. Test results confirm the feasibility of our approach. In addition, the stress testing results show that even unoptimized implementations of our algorithms do not affect the performance of the OpenFlow controller significantly.
     1335</li>
     1336<br>
     1337
     1338
     1339
     1340<li>
     1341<b>Jofre, Jordi and Velayos, Celia and Landi, Giada and Giertych, Micha&#x0142; and Hume, Alastair C. and Francis, Gareth and Vico Oton, Albert</b>
     1342, &quot;Federation of the BonFIRE multi-cloud infrastructure with networking facilities.&quot;
     1343Computer Networks,
     13442014.
     1345doi:10.1016/j.bjp.2013.11.012.
     1346<a href="http://dx.doi.org/10.1016/j.bjp.2013.11.012">http://dx.doi.org/10.1016/j.bjp.2013.11.012</a>
     1347<br><br><b>Abstract: </b>Network performance in terms of throughput, latency, packet loss or jitter significantly influences user's quality of experience of cloud applications. Network services impact on cloud applications performance and this impact is even more significant when the cloud infrastructure spreads over different administrative domains, such as in a federated cloud or hybrid-cloud scenarios. Given this strong coupling between cloud application performance and network performance there is great value to be gained by supporting advanced controlled networking functionalities between distributed cloud infrastructures. These functionalities would be useful to the Future Internet (FI) experimentation community as well as future production clouds. This paper describes an architecture and a set of procedures to interconnect a multi-cloud environment with advanced facilities for controlled networking. This integration allows the provisioning of customized network functions and services in support of experiments running in a multi-cloud test-bed. The possibility to control the network connectivity is a key feature to provide better performance for the experimenters' cloud applications. We focus on the details of federating three advanced networking facilities with the BonFIRE multi-cloud environment. These three networking facilities are: FEDERICA, which supports controlled routing; G&#x45;&#x0301;ANT's Bandwidth-on-Demand service and OFELIA that uses OpenFlow to provide Software Defined Network functionalities. The interconnections with FEDERICA and G&#x45;&#x0301;ANT are already active, while OFELIA is envisaged as future work for a third facility to interconnect.
     1348</li>
     1349<br>
     1350
     1351
     1352
     1353<li>
     1354<b>Jourjon, Guillaume and Marquez-Barja, Johann M. and Rakotoarivelo, Thierry and Mikroyannidis, Alexander and Lampropoulos, Kostas and Denazis, Spyros and Tranoris, Christos and Pareit, Daan and Domingue, John and DaSilva, Luiz A. and Ott, Max</b>
     1355, &quot;FORGE Toolkit: Leveraging Distributed Systems in eLearning Platforms.&quot;
     1356IEEE,
     13572015.
     1358doi:10.1109/tetc.2015.2511454.
     1359<a href="http://dx.doi.org/10.1109/tetc.2015.2511454">http://dx.doi.org/10.1109/tetc.2015.2511454</a>
     1360<br><br><b>Abstract: </b>While more and more services become virtualised and always accessible in our society, laboratories supporting Computer Science (CS) lectures have mainly remained offline and class-based. This apparent abnormality is due to several limiting factors, discussed in the literature, such as the high cost of deploying and maintaining computer network testbeds and the lack of standardisation for the presentation of eLearning platforms. In this paper, we present the FORGE toolkit, which leverages experimentation facilities currently deployed in international initiatives for the development of e-learning materials. Thus, we solve the institutional challenge mentioned in the ACM/IEEE 2013 CS curricula concerning the access and maintenance of specialised and heterogeneous hardware thanks to a seamless integration with the networking testbed community. Moreover, this project builds an ecosystem where teaching and educational materials, tools and experiments are available under open scheme and policies. We demonstrate how it already meets most of the requirements from the Network and Communication component of CS 2013 and some of the labs of the Cisco academy. Finally, we present experience reports illustrating the potential benefits of this framework based on first deployments in four post-graduate courses in prestigious institutions around the world.
     1361</li>
     1362<br>
     1363
     1364
     1365
     1366<li>
     1367<b>Ju, Xi and Zhang, Hongwei and Zeng, Wenjie and Sridharan, Mukundan and Li, Jing and Arora, Anish and Ramnath, Rajiv and Xin, Yufeng</b>
     1368, &quot;LENS: resource specification for wireless sensor network experimentation infrastructures.&quot;
     1369Proceedings of the 6th ACM international workshop on Wireless network testbeds, experimental evaluation and characterization, Las Vegas, Nevada, USA, ACM, New York, NY, USA,
     13702011.
     1371doi:10.1145/2030718.2030727.
     1372<a href="http://dx.doi.org/10.1145/2030718.2030727">http://dx.doi.org/10.1145/2030718.2030727</a>
     1373<br><br><b>Abstract: </b>As a first step towards predictable, repeatable WSN experimentation, we propose the resource specification language LENS (a.k.a. Language for Embedded Networked Sensing) for WSN experimentation infrastructures. Using the Resource Description Framework (RDF) and the Web Ontology Language (OWL), LENS defines a semantic ontology for WSN resources; LENS enables explicit control and measurement of uncertainty factors, and it enables reasoning about the relationships between WSN resources. Focusing on basic concepts of WSNs, LENS supports resource specification in a wide range of WSN experimentation infrastructures, and it is extensible to support potentially unforeseen technologies. LENS is also compatible with specification languages for other network resources such as optical networks. As a part of the NSF GENI initiative, we have implemented LENS in the KanseiGenie control framework, and LENS has been actively used to support experimentation in the federated WSN infrastructure involving Kansei and NetEye. Enabling reasoning about uncertainty factors in experimentation, LENS is expected to serve as a basis for developing methodologies and tools for predictable, repeatable WSN experimentation.
     1374</li>
     1375<br>
     1376
     1377
     1378
     1379<li>
     1380<b>Juluri, Parikshit</b>
     1381, &quot;Measurement And Improvement of Quality-of-Experience For Online Video Streaming Services (Doctoral dissertation).&quot;
     1382
     13832015.
     1384
     1385<a href="https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/46696/JuluriMeaImpQua.pdf?sequence=1&#x0026;&#x0023;38;isAllowed=y">https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/46696/JuluriMeaImpQua.pdf?sequence=1&#x0026;&#x0023;38;isAllowed=y</a>
     1386<br><br><b>Abstract: </b>HTTP based online video streaming services have been consistently dominating the online traffic for the past few years. Measuring and improving the performance of these services is an important challenge. Traditional Quality-of-Service (QoS) metrics such as packet loss, jitter and delay which were used for networked services are not easily understood by the users. Instead, Quality-of-Experience (QoE) metrics which capture the overall satisfaction are more suitable for measuring the quality as perceived by the users. However, these QoE metrics have not yet been standardized and their measurement and improvement poses unique challenges. In this work we first present a comprehensive survey of the different set of QoE metrics and the measurement methodologies suitable for HTTP based online video streaming services. We then present our active QoE measurement tool Pytomo that measures the QoE of YouTube videos. A case study on the measurement of QoE of YouTube videos when accessed by residential users from three different Internet Service Providers (ISP) in a metropolitan area is discussed. This is the first work that has collected QoE data from actual residential users using active measurements for YouTube videos. Based on these measurements we were able to study and compare the QoE of YouTube videos across multiple ISPs. We also were able to correlate the QoE observed with the server clusters used for the different users. Based on this correlation we were able to identify the server clusters that were experiencing diminished QoE. DynamicAdaptive Streaming overHTTP (DASH) is an HTTP based video streaming that enables the video players to adapt the video quality based on the network conditions. We next present a rate adaptation algorithm that improves the QoE of DASH video streaming services that selects the most optimum video quality. With DASH the video server hosts multiple representation of the same video and each representation is divided into small segments of constant playback duration. The DASH player downloads the appropriate representation based on the network conditions, thus, adapting the video quality to match the conditions. Currently deployed Adaptive Bitrate (ABR) algorithms use throughput and buffer occupancy to predict segment fetch times. These algorithms assume that the segments are of equal size. However, due to the encoding schemes employed this assumption does not hold. In order to overcome these limitations, we propose a novel Segment Aware Rate Adaptation algorithm (SARA) that leverages the knowledge of the segment size variations to improve the prediction of segment fetch times. Using an emulated player in a geographically distributed virtual network setup, we compare the performance of SARA with existing ABR algorithms. We demonstrate that SARA helps to improve the QoE of the DASH video streaming with improved convergence time, better bitrate switching performance and better video quality. We also show that unlike the existing adaptation schemes, SARA provides a consistent QoE irrespective of the segment size distributions.
     1387</li>
     1388<br>
     1389
     1390
     1391
     1392<li>
     1393<b>Juluri, Parikshit and Tamarapalli, Venkatesh and Medhi, Deep</b>
     1394, &quot;SARA: Segment aware rate adaptation algorithm for dynamic adaptive streaming over HTTP.&quot;
     1395Communication Workshop (ICCW), 2015 IEEE International Conference on, IEEE,
     13962015.
     1397doi:10.1109/iccw.2015.7247436.
     1398<a href="http://dx.doi.org/10.1109/iccw.2015.7247436">http://dx.doi.org/10.1109/iccw.2015.7247436</a>
     1399<br><br><b>Abstract: </b>Dynamic adaptive HTTP (DASH) based streaming is steadily becoming the most popular online video streaming technique. DASH streaming provides seamless playback by adapting the video quality to the network conditions during the video playback. A DASH server supports adaptive streaming by hosting multiple representations of the video and each representation is divided into small segments of equal playback duration. At the client end, the video player uses an adaptive bitrate selection (ABR) algorithm to decide the bitrate to be selected for each segment depending on the current network conditions. Currently, proposed ABR algorithms ignore the fact that the segment sizes significantly vary for a given video bitrate. Due to this, even though an ABR algorithm is able to measure the network bandwidth, it may fail to predict the time to download the next segment In this paper, we propose a segment-aware rate adaptation (SARA) algorithm that considers the segment size variation in addition to the estimated path bandwidth and the current buffer occupancy to accurately predict the time required to download the next segment We also developed an open source Python based emulated DASH video player, that was used to compare the performance of SARA and a basic ABR. Our results show that SARA provides a significant gain over the basic algorithm in the video quality delivered, without noticeably impacting the video switching rates.
     1400</li>
     1401<br>
     1402
     1403
     1404
     1405<li>
     1406<b>Kanada, Yasusi and Tarui, Toshiaki</b>
     1407, &quot;Federation-less federation of ProtoGENI and VNode platforms.&quot;
     1408Information Networking (ICOIN), 2015 International Conference on, IEEE,
     14092015.
     1410doi:10.1109/icoin.2015.7057895.
     1411<a href="http://dx.doi.org/10.1109/icoin.2015.7057895">http://dx.doi.org/10.1109/icoin.2015.7057895</a>
     1412<br><br><b>Abstract: </b>Our previous work enabled ” federation-less federation”, which means a federation of multiple network-virtualization platforms that do not support federation functions, and applied this method to a homogeneous federation of platforms called the ” VNode” infrastructures. In this study, this method was applied to a heterogeneous federation of the ProtoGENI and the ” VNode”. We intended to federate these platforms through a single management interface. However, the federation architecture of GENI, which is called the slice-based federation architecture (SFA), cannot be used for single-interface federation but we could not modify the ProtoGENI platform to enable it. Therefore, a method for applying federation-less-federation to ProtoGENI was developed. It enabled federation of these platforms by adding several nodes but without modifying preexisting platforms. This method was applied to federation of the ProtoGENI platform at the University of Utah and two VNode infrastructures in Japan, the slice creation and deletion time was measured and evaluated to be acceptable. Although this federation-less-federation implementation still has several minor problems, it was proved to be useful for experiments and demonstrations.
     1413</li>
     1414<br>
     1415
     1416
     1417
     1418<li>
     1419<b>Kangarlou, A. and Xu, Dongyan and Kozat, U. C. and Padala, P. and Lantz, B. and Igarashi, K.</b>
     1420, &quot;In-network live snapshot service for recovering virtual infrastructures.&quot;
     1421Network, IEEE, IEEE,
     14222011.
     1423doi:10.1109/mnet.2011.5958003.
     1424<a href="http://dx.doi.org/10.1109/mnet.2011.5958003">http://dx.doi.org/10.1109/mnet.2011.5958003</a>
     1425<br><br><b>Abstract: </b>Infrastructure as a Service (IaaS) has become an increasingly popular type of service for both private and public clouds. The virtual infrastructures that enable IaaS support multitenancy by multiplexing the computational resources of data centers and result in substantial reductions in operational costs. Since hardware and software failures occur on a routine basis in large-scale systems, it is imperative for cloud providers to offer various failure recovery options for distributed services hosted on such infrastructures. In this article we present GENI-VIOLIN, a new cloud capability that can checkpoint a stateful distributed service while incurring very low overhead. The unique aspect of GENI-VIOLIN compared to previous work is that GENI-VIOLIN exploits programmable OpenFlow switches to provide checkpointing services in the network, thereby requiring minimal changes to the end host virtualization framework. We have developed a prototype of GENI-VIOLIN using the GENI infrastructure, and have demonstrated GENI-VIOLIN's checkpoint and restore capability across multiple GENI sites.
     1426</li>
     1427<br>
     1428
     1429
     1430
     1431<li>
     1432<b>Katz-Bassett, Ethan and Choffnes, David R. and Cunha, &#x49;&#x0301;talo and Scott, Colin and Anderson, Thomas and Krishnamurthy, Arvind</b>
     1433, &quot;Machiavellian Routing: Improving Internet Availability with BGP Poisoning.&quot;
     1434Proceedings of the 10th ACM Workshop on Hot Topics in Networks, Cambridge, Massachusetts, ACM, New York, NY, USA,
     14352011.
     1436doi:10.1145/2070562.2070573.
     1437<a href="http://dx.doi.org/10.1145/2070562.2070573">http://dx.doi.org/10.1145/2070562.2070573</a>
     1438<br><br><b>Abstract: </b>We propose a new approach to mitigate disruptions of Internet connectivity. The Internet was designed to always find a route if there is a policy-compliant path; however, in many cases, connectivity is disrupted despite the existence of an underlying valid path. The research community has done considerable work on this problem, much of it focused on short-term outages that occur during route convergence. There has been less progress on addressing avoidable long-lasting outages. Our measurements show that long-lasting events contribute significantly to overall unavailability. To address these long-term problems, we develop a system, Machiavellian routing, for automatic failure remediation, centered around the use of BGP poisoning. With poisoning, an edge network can cause other networks to send traffic to it via paths that avoid a problem in a particular transit ISP. We describe the key challenges to using poisoning to improve Internet connectivity, and we develop a set of techniques to use it predictably, accurately, and effectively.
     1439</li>
     1440<br>
     1441
     1442
     1443
     1444<li>
     1445<b>Katz-Bassett, Ethan and Scott, Colin and Choffnes, David R. and Cunha, &#x49;&#x0301;talo and Valancius, Vytautas and Feamster, Nick and Madhyastha, Harsha V. and Anderson, Thomas and Krishnamurthy, Arvind</b>
     1446, &quot;LIFEGUARD: Practical Repair of Persistent Route Failures.&quot;
     1447Proceedings of the ACM SIGCOMM 2012 conference, ACM, New York, NY, USA,
     14482012.
     1449doi:10.1145/2377677.2377756.
     1450<a href="http://dx.doi.org/10.1145/2377677.2377756">http://dx.doi.org/10.1145/2377677.2377756</a>
     1451<br><br><b>Abstract: </b>The Internet was designed to always find a route if there is a policy-compliant path. However, in many cases, connectivity is disrupted despite the existence of an underlying valid path. The research community has focused on short-term outages that occur during route convergence. There has been less progress on addressing avoidable long-lasting outages. Our measurements show that long-lasting events contribute significantly to overall unavailability. To address these problems, we develop LIFEGUARD, a system for automatic failure localization and remediation. LIFEGUARD uses active measurements and a historical path atlas to locate faults, even in the presence of asymmetric paths and failures. Given the ability to locate faults, we argue that the Internet protocols should allow edge ISPs to steer traffic to them around failures, without requiring the involvement of the network causing the failure. Although the Internet does not explicitly support this functionality today, we show how to approximate it using carefully crafted BGP messages. LIFEGUARD employs a set of techniques to reroute around failures with low impact on working routes. Deploying LIFEGUARD on the Internet, we find that it can effectively route traffic around an AS without causing widespread disruption.
     1452</li>
     1453<br>
     1454
     1455
     1456
     1457<li>
     1458<b>Khurshid, Ahmed and Zhou, Wenxuan and Caesar, Matthew and Godfrey, P. Brighten</b>
     1459, &quot;VeriFlow: verifying network-wide invariants in real time.&quot;
     1460Proceedings of the first workshop on Hot topics in software defined networks, Helsinki, Finland, ACM, New York, NY, USA,
     14612012.
     1462doi:10.1145/2342441.2342452.
     1463<a href="http://doi.acm.org/10.1145/2342441.2342452">http://doi.acm.org/10.1145/2342441.2342452</a>
     1464<br><br><b>Abstract: </b>Networks are complex and prone to bugs. Existing tools that check configuration files and data-plane state operate offline at timescales of seconds to hours, and cannot detect or prevent bugs as they arise. Is it possible to check network-wide invariants in real time, as the network state evolves? The key challenge here is to achieve extremely low latency during the checks so that network performance is not affected. In this paper, we present a preliminary design, VeriFlow, which suggests that this goal is achievable. VeriFlow is a layer between a software-defined networking controller and network devices that checks for network-wide invariant violations dynamically as each forwarding rule is inserted. Based on an implementation using a Mininet OpenFlow network and Route Views trace data, we find that VeriFlow can perform rigorous checking within hundreds of microseconds per rule insertion.
     1465</li>
     1466<br>
     1467
     1468
     1469
     1470<li>
     1471<b>Kim, Dae Y. and Mathy, Laurent and Campanella, Mauro and Summerhill, Rick and Williams, James and Shimojo, Shinji and Kitamura, Yasuichi and Otsuki, Hideaki</b>
     1472, &quot;Future Internet: Challenges in Virtualization and Federation.&quot;
     14732009 Fifth Advanced International Conference on Telecommunications, Venice/Mestre, Italy, IEEE,
     14742009.
     1475doi:10.1109/AICT.2009.8.
     1476<a href="http://dx.doi.org/10.1109/AICT.2009.8">http://dx.doi.org/10.1109/AICT.2009.8</a>
     1477<br><br><b>Abstract: </b>Future Internet is a clean-slate research activity in the quest of new networking technologies to overcome the limits of the current Internet. In its experimental research, virtualization and federation are emerging as essential features, especially in the construction and operation of the testbeds. Moreover, they are believed to sustain as the fundamental features of the Future Internet itself. Visions and experiences on virtualization and federation are given by leading experts from US, EU, and Asia.
     1478</li>
     1479<br>
     1480
     1481
     1482
     1483<li>
     1484<b>Kim, Dongkyun and Kim, Joobum and Wang, Gicheol and Park, Jin-Hyung and Kim, Seung-Hae</b>
     1485, &quot;K-GENI testbed deployment and federated meta operations experiment over GENI and KREONET.&quot;
     1486Computer Networks,
     14872014.
     1488doi:10.1016/j.bjp.2013.11.016.
     1489<a href="http://dx.doi.org/10.1016/j.bjp.2013.11.016">http://dx.doi.org/10.1016/j.bjp.2013.11.016</a>
     1490<br><br><b>Abstract: </b>The classical Internet has confronted many drawbacks in terms of network security, scalability, and performance, although it has strongly influenced the development and evolution of diverse network technologies, applications, and services. Therefore, new innovative research on the Future Internet has been performed to resolve the inherent weaknesses of the traditional Internet, which, in turn, requires new at-scale network testbeds and research infrastructure for large-scale experiments. In this context, K-GENI has been developed as an international programmable Future Internet testbed in the GENI spiral-2 program, and it has been operational between the USA (GENI) and Korea (KREONET) since 2010. The K-GENI testbed and the related collaborative efforts will be introduced with two major topics in this paper: (1) the design and deployment of the K-GENI testbed and (2) the federated meta operations between the K-GENI and GENI testbeds. Regarding the second topic in particular, we will describe how meta operations are federated across K-GENI between GMOC (GENI Meta Operations Center) and DvNOC (Distributed virtual Network Operations Center on KREONET/K-GENI), which is the first trial of an international experiment on the federated network operations over GENI.
     1491</li>
     1492<br>
     1493
     1494
     1495
     1496<li>
     1497<b>Kim, Hyunjun and Lee, Sungwon</b>
     1498, &quot;FiRST Cloud Aggregate Manager development over FiRST: Future Internet testbed.&quot;
     1499The International Conference on Information Network 2012, Bali, Indonesia, IEEE,
     15002012.
     1501doi:10.1109/ICOIN.2012.6164436.
     1502<a href="http://dx.doi.org/10.1109/ICOIN.2012.6164436">http://dx.doi.org/10.1109/ICOIN.2012.6164436</a>
     1503<br><br><b>Abstract: </b>FiRST (Future Internet Research for Sustainable Test-bed) is the future internet platform development project being performed in Korea. The goal of the project is to create the virtualized and dynamic service creation environments over future internet networks; it is an experimental project to realize future innovative service ideas over real network environments. Among this, cloud computing is the key enabler to control and allocate virtualized network resources (such as CPU, storage, and virtualized network configuration) for the requested services. However, researches on interworking between future internet and cloud computing is in initial phase. In this paper, we propose the FiRST Cloud Aggregate Manager (AM) based on GENI (Global Environment for Network Innovation) AM Application Programming Interface (API) for the federation between future internet test-bed and open source OpenStack cloud computing platform. After that, we propose the zero-client service for mobile cloud management. In order to control the zero-client service, we develop Cloud Mobility Client/Server. And, we validate and verified our FiRST Cloud AM and zero-client service by developing experimental test-bed. Through this test-bed, we confirm that the proposed FiRST Cloud AM and zero-client service efficiently interworks with future internet control plane framework by using GENI Control Framework (GCF) tools.
     1504</li>
     1505<br>
     1506
     1507
     1508
     1509<li>
     1510<b>Kline, Donald and Quan, John</b>
     1511, &quot;Attribute description service for large-scale networks.&quot;
     1512Proceedings of the 2nd international conference on Human centered design, Orlando, FL, USA, Springer-Verlag, Berlin, Heidelberg,
     15132011.
     1514doi:10.1007/978-3-642-21753-1&#x005F;58.
     1515<a href="http://portal.acm.org/citation.cfm?id=2021672.2021735">http://portal.acm.org/citation.cfm?id=2021672.2021735</a>
     1516<br><br><b>Abstract: </b>An analysis of requesting resources from large-scale networks reveals a fundamental challenge. As the network grows, more and more resources become available, and so finding resources that fit experimental test criteria becomes difficult and time consuming. For example, the National Science Foundation sponsors GENI--an experimental network with a goal to gain enough resources to model the Internet at scale. Currently, GENI contains relatively few contributed resources donated from businesses and academia, and so matching resources to tests is rather simple. However, experimenters plan to conduct network experiments that are very complex and difficult to accurately model by using the vast numbers of resources expected in GENI. When GENI reaches its final state, finding the right resources that fit experimental test criteria out of many thousands of donated resources may be as difficult as conducting the experiment itself. This dilemma underscores the importance of establishing an attribute description service that promotes a standardized language for all interactions between the end users and the large-scale network.
     1517</li>
     1518<br>
     1519
     1520
     1521
     1522<li>
     1523<b>Kobayashi, Masayoshi and Seetharaman, Srini and Parulkar, Guru and Appenzeller, Guido and Little, Joseph and van Reijendam, Johan and Weissmann, Paul and McKeown, Nick</b>
     1524, &quot;Maturing of OpenFlow and Software-defined Networking through deployments.&quot;
     1525Computer Networks,
     15262014.
     1527doi:10.1016/j.bjp.2013.10.011.
     1528<a href="http://dx.doi.org/10.1016/j.bjp.2013.10.011">http://dx.doi.org/10.1016/j.bjp.2013.10.011</a>
     1529<br><br><b>Abstract: </b>Software-defined Networking (SDN) has emerged as a new paradigm of networking that enables network operators, owners, vendors, and even third parties to innovate and create new capabilities at a faster pace. The SDN paradigm shows potential for all domains of use, including data centers, cellular providers, service providers, enterprises, and homes. Over a three-year period, we deployed SDN technology at our campus and at several other campuses nation-wide with the help of partners. These deployments included the first-ever SDN prototype in a lab for a (small) global deployment. The four-phased deployments and demonstration of new networking capabilities enabled by SDN played an important role in maturing SDN and its ecosystem. We share our experiences and lessons learned that have to do with demonstration of SDN's potential; its influence on successive versions of OpenFlow specification; evolution of SDN architecture; performance of SDN and various components; and growing the ecosystem.
     1530</li>
     1531<br>
     1532
     1533
     1534
     1535<li>
     1536<b>Krishnappa, D. K. and Irwin, D. and Lyons, E. and Zink, M.</b>
     1537, &quot;CloudCast: Cloud Computing for Short-Term Weather Forecasts.&quot;
     1538Computing in Science &#x0026; Engineering, IEEE,
     15392013.
     1540doi:10.1109/mcse.2013.43.
     1541<a href="http://dx.doi.org/10.1109/mcse.2013.43">http://dx.doi.org/10.1109/mcse.2013.43</a>
     1542<br><br><b>Abstract: </b>CloudCast provides personalized short-term weather forecasts to clients based on their current location using cloud services, generating accurate forecasts tens of minutes in the future for small areas. Results show that it takes less than two minutes from the start of data sampling to deliver a 15-minute forecast to a client.
     1543</li>
     1544<br>
     1545
     1546
     1547
     1548<li>
     1549<b>Krishnappa, Dilip K. and Lyons, Eric and Irwin, David and Zink, Michael</b>
     1550, &quot;Performance of GENI Cloud Testbeds for Real Time Scientific Application.&quot;
     1551First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     15522012.
     1553
     1554
     1555<br><br><b>Abstract: </b>Dedicating high end servers for short-term execution of scientific applications such as weather forecasting wastes resources. Cloud platforms IaaS model seems well suited for applications which are executed on an irregular basis and for short duration. In this paper, we evaluate the performance of research testbed cloud platforms such as GENICloud and ORCA cloud clusters for our real-time scientific application of short-term weather forecasting called Nowcasting. In this paper, we evaluate the network capabilities of these research cloud testbeds for our real-time application of weather forecasting. In addition, we evaluate the computation time of executing Nowcasting on each cloud platform for weather data collected from real weather events. We also evaluate the total time taken to generate and transmit short-term forecast images to end users with live data from our own radar on campus. We also compare the performance of each of these clusters for Nowcasting with commercial cloud services such as Amazon's EC2. The results obtained from our measurement show that cloud testbeds are suitable for real-time application experiments to be carried out on a cloud platform.
     1556</li>
     1557<br>
     1558
     1559<li>
     1560<b>Krishnappa, Dilip K. and Lyons, Eric and Irwin, David and Zink, Michael</b>
     1561, &quot;Network capabilities of cloud services for a real time scientific application.&quot;
     156237th Annual IEEE Conference on Local Computer Networks, Clearwater Beach, FL, USA, IEEE,
     15632012.
     1564doi:10.1109/lcn.2012.6423665.
     1565<a href="http://dx.doi.org/10.1109/lcn.2012.6423665">http://dx.doi.org/10.1109/lcn.2012.6423665</a>
     1566<br><br><b>Abstract: </b>Dedicating high-end servers for executing scientific applications that run intermittently, such as severe weather detection or generalized weather forecasting, wastes resources. While the Infrastructure-as-a-Service (IaaS) model used by today's cloud platforms is well-suited for the bursty computational demands of these applications, it is unclear if the network capabilities of today's cloud platforms are sufficient. In this paper, we analyze the networking capabilities of multiple commercial (Amazon's EC2 and Rackspace) and research (GENICloud and ExoGENI cloud) platforms in the context of a Nowcasting application, a forecasting algorithm for highly accurate, near-term, e.g., 5-20 minutes, weather predictions. The application has both computational and network requirements. While it executes rarely, whenever severe weather approaches, it benefits from an IaaS model; However, since its results are time-critical, enough bandwidth must be available to transmit radar data to cloud platforms before it becomes stale. We conduct network capacity measurements between radar sites and cloud platforms throughout the country. Our results indicate that ExoGENI cloud performs the best for both serial and parallel data transfer with an average throughput of 110.22 Mbps and 17.2 Mbps, respectively. We also found that the cloud services perform better in the distributed data transfer case, where a subset of nodes transmit data in parallel to a cloud instance. Ultimately, we conclude that commercial and research clouds are capable of providing sufficient bandwidth for our real-time Nowcasting application.
     1567</li>
     1568<br>
     1569
     1570
     1571
     1572<li>
     1573<b>Kuai, Meng and Hong, Xiaoyan and Flores, R. R.</b>
     1574, &quot;Evaluating Interest Broadcast in Vehicular Named Data Networking.&quot;
     1575Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     15762014.
     1577doi:10.1109/gree.2014.23.
     1578<a href="http://dx.doi.org/10.1109/gree.2014.23">http://dx.doi.org/10.1109/gree.2014.23</a>
     1579<br><br><b>Abstract: </b>Vehicular Ad-hoc Networks (VANETs) are expected to provide assistance to various applications, such as accident notification and emergency announcement. Named Data Networking (NDN) has been recognized as a more suitable architecture than TCP/IP for application in VANETs due to its ability to handle high mobility and intermittent connectivity. The Vehicular NDN (V-NDN) has further made special architectural modifications for VANETs. However, V-NDN can be challenged in its extensive use of broadcast in dense network situations. For example, broadcasting of interest packets could lead to more collisions. In this study, we explore the broadcast performance of V-NDN using the ORBIT testbed. Our experimental results show that VNDN suffers an increased loss ratio in dense network scenarios because of Wifi broadcast collision, and it is important to find a suitable range of values to be distributed by the collision avoidance timer before transmission.
     1580</li>
     1581<br>
     1582
     1583
     1584
     1585<li>
     1586<b>Lara, Adrian</b>
     1587, &quot;Using Software-Defined Networking to Improve Campus, Transport and Future Internet Architectures (Doctoral dissertation).&quot;
     1588
     15892015.
     1590
     1591<a href="http://digitalcommons.unl.edu/computerscidiss/93/">http://digitalcommons.unl.edu/computerscidiss/93/</a>
     1592<br><br><b>Abstract: </b>Software-defined Networking (SDN) promises to redefine the future of networking. Indeed, SDN-based networks have unique capabilities such as centralized control, flow abstraction, dynamic updating of forwarding rules and software-based traffic analysis. SDN-based networks decouple the data plane from the control plane, migrating the latter to a software controller. By adding a software layer between network devices and applications, features such as network virtualization and automated management are simpler to achieve. In this dissertation, we show how SDN-based deployments simplify network management at multiple scales such as campus and transport networks, as well as future Internet architectures. First, we propose OpenSec, an SDN-based security framework that allows network operators to implement security policies in campus networks. Second, we propose the eXtensible Traffic Engineering Framework (XTEF) to enable application-driven traffic engineering and provision transport network resources using on-demand Wavelength Division Multiplexing (WDM) tunnels. Third, we demonstrate how SDN can be used to dynamically create intra-domain cut-through switching tunnels to bypass the routing layer in MobilityFirst. Finally, we propose how to extend the cut-through capabilities to inter-domain routing in MobilityFirst. In our work, we run experiments on the GENI testbed (Global Environment for Network Innovations), the ORBIT (Open-Access Research Testbed for Next-Generation Wireless Networks) and Mininet. The results show that SDN can be used to simplify policy-based network management, virtualize an entire WAN as a single switch, create Wavelength Division Multiplexing (WDM) tunnels on demand and create inter-domain tunnels using techniques that scale better than traditional distributed methods.
     1593</li>
     1594<br>
     1595
     1596
     1597
     1598<li>
     1599<b>Lara, Adrian and Ramamurthy, Byrav and Nagaraja, Kiran and Krishnamoorthy, Aravind and Raychaudhuri, Dipankar</b>
     1600, &quot;Using OpenFlow to provide cut-through switching in MobilityFirst.&quot;
     1601Photonic Network Communications, Springer US,
     16022014.
     1603doi:10.1007/s11107-014-0461-3.
     1604<a href="http://dx.doi.org/10.1007/s11107-014-0461-3">http://dx.doi.org/10.1007/s11107-014-0461-3</a>
     1605<br><br><b>Abstract: </b>Mobile devices are expected to become the Internet's predominant technology. Current protocols such as TCP/IP were not originally designed with mobility as a key consideration, and therefore underperform under challenging mobile and wireless conditions. MobilityFirst, a clean slate architecture proposal, embraces several key concepts centered around secure identifiers that inherently support mobility and trustworthiness as key requirements of the network architecture. This includes a hop-by-hop segmented data transport based on a globally unique identifier. This allows late and dynamic rebinding of end-point addresses to support mobility. While this provides critical gains in wireless segments, some overheads are incurred even in stable segments such as in the core. Bypassing routing-layer decisions in these cases, with lower layer cut-through forwarding, can improve said gains. In this work, we introduce a general bypass capability within the MobilityFirst architecture that provides better performance and enables both individual and aggregate flow-level traffic control. Furthermore, we present an OpenFlow-based proof-of-concept implementation of the bypass function using layer 2 VLAN tagging. We run experiments on the ORBIT and Global Environment for Network Innovations (GENI) testbeds to evaluate the performance and scalability of the solution. By implementing the bypass functionality, we are able to significantly reduce the number of messages processed by the controller as well as the number of flow rules that need to be pushed into the switches.
     1606</li>
     1607<br>
     1608
     1609
     1610
     1611<li>
     1612<b>Lauer, Gregory and Irwin, Ryan and Kappler, Chris and Nishioka, Itaru</b>
     1613, &quot;Distributed Resource Control Using Shadowed Subgraphs.&quot;
     1614Proceedings of the Ninth ACM Conference on Emerging Networking Experiments and Technologies, Santa Barbara, California, USA, ACM, New York, NY, USA,
     16152013.
     1616doi:10.1145/2535372.2535410.
     1617<a href="http://dx.doi.org/10.1145/2535372.2535410">http://dx.doi.org/10.1145/2535372.2535410</a>
     1618<br><br><b>Abstract: </b>As software defined networks (SDN) grow in size and in number, the problem of coordinating the actions of multiple SDN controllers will grow in importance. In this paper, we propose a way of organizing SDN control based on coordinated subgraph shadowing. Graphs are a natural way to think about and describe SDN activity. Subgraphs provide a means to share a subset of a network's resources. Shadowing provides a means to dynamically update shared subgraphs. Leveraging advances in graph databases and our shadowing messaging technique, we discuss our implementation of a multi-domain virtual private network (VPN) using multi-protocol label switching (MPLS).
     1619</li>
     1620<br>
     1621
     1622
     1623
     1624<li>
     1625<b>Lee, Jae W.</b>
     1626, &quot;Towards a Common System Architecture for Dynamically Deploying Network Services in Routers and End Hosts (Doctoral dissertation).&quot;
     1627
     16282012.
     1629
     1630<a href="http://academiccommons.columbia.edu/download/fedora&#x005F;content/download/ac:147210/CONTENT/Lee&#x005F;columbia&#x005F;0054D&#x005F;10773.pdf">http://academiccommons.columbia.edu/download/fedora&#x005F;content/download/ac:147210/CONTENT/Lee&#x005F;columbia&#x005F;0054D&#x005F;10773.pdf</a>
     1631<br><br><b>Abstract: </b>The architectural simplicity of the core Internet is a double-edged sword. On the one hand, its agnostic nature paved the way for endless innovations of end-to-end applications. On the other hand, the inherent limitation of this simplicity makes it difficult to add new functions to the network core itself. This is exacerbated by the conservative tendency of commercial entities to &#x6c;&#x0308;eave well-enough alone&#x2c;&#x0308; leading to the current situation often referred to as the ossification of the Internet. For decades, there has been practically no new functionality that has been added to the core Internet on a large scale. This thesis explores the possibility of enabling in-network services towards the goal of overcoming the ossification of the Internet. Our ultimate goal is to provide a common run-time environment supported by all Internet nodes and a wide-area deployment mechanism, so that network services can be freely installed, removed, and migrated among Internet nodes of all kinds–from a backbone router to a set-top box at home. In that vision of a future Internet, there is little difference between servers and routers for the purpose of running network services. Services can run anywhere on the Internet. Application service providers will have the freedom to choose the best place to run their code. This thesis presents NetServ, our first step to realize the vision of network services running anywhere on the Internet. NetServ is a node architecture for dynamically deploying in-network services on edge routers. Network functions and applications are implemented as software modules which can be deployed at any NetServ-enabled node on the Internet, subject to policy restrictions. The NetServ framework provides a common execution environment for service modules and the ability to dynamically install and remove the services without restarting the nodes. There are many challenges in designing such a system. The main contribution of this thesis lies in meeting those challenges. First, we recognize that the primary impetus for adopting new technologies is economics. To address the challenge of providing economic incentives for enabling in-network services, we demonstrate how NetServ can facilitate an economic alliance between content providers and ISPs. Using NetServ, content providers and the ISPs operating at the network edge (aka eyeball ISPs) can enter into a mutually beneficial economic relationship. ISPs make their NetServ-enabled edge routers available for hosting content providers' applications and contents. Content providers can operate closer to end users by deploying code modules on NetServ-enabled edge routers. We make our case by presenting NetServ applications which represent four concrete use cases. Second, our node architecture must support both traditional server applications and in-network packet processing applications since content providers' applications running on ISPs' routers will combine the traits of both. To address this challenge, NetServ framework can host a packet processing module that sits in the data path, a server module that uses the TCP/IP stack in the traditional way, or a combined module that does both. NetServ provides a unified runtime environment between routers and servers, taking us a step closer to the vision of the unified runtime available on all Internet nodes. Third, we must provide a fast and streamlined deployment mechanism. Content providers should be able to deploy their applications at any NetServ-enabled edge router on the Inter- net, given that they have proper authorizations. Moreover, in some application scenarios, content providers may not know the exact locations of the target routers. Content providers need a way to send a message to install or remove an application module towards a network destination, and have the NetServ-enabled routers located in the path catch and act on the message. To address this challenge, we adopted on-path signaling as the deployment mechanism for NetServ. A NetServ signaling message is sent in an IP packet towards a destination. The packet gets forwarded by IP routers as usual, but when it transits a NetServ-enabled router, the message gets intercepted and passed to the NetServ control layer. Fourth, a NetServ-enabled router must support the concurrent executions of multiple without restarting the nodes. There are many challenges in designing such a system. The main contribution of this thesis lies in meeting those challenges. First, we recognize that the primary impetus for adopting new technologies is economics. To address the challenge of providing economic incentives for enabling in-network services, we demonstrate how NetServ can facilitate an economic alliance between content providers and ISPs. Using NetServ, content providers and the ISPs operating at the network edge (aka eyeball ISPs) can enter into a mutually beneficial economic relationship. ISPs make their NetServ-enabled edge routers available for hosting content providers' applications and contents. Content providers can operate closer to end users by deploying code modules on NetServ-enabled edge routers. We make our case by presenting NetServ applications which represent four concrete use cases. Second, our node architecture must support both traditional server applications and in-network packet processing applications since content providers' applications running on ISPs' routers will combine the traits of both. To address this challenge, NetServ framework can host a packet processing module that sits in the data path, a server module that uses the TCP/IP stack in the traditional way, or a combined module that does both. NetServ provides a unified runtime environment between routers and servers, taking us a step closer to the vision of the unified runtime available on all Internet nodes. Third, we must provide a fast and streamlined deployment mechanism. Content providers should be able to deploy their applications at any NetServ-enabled edge router on the Internet, given that they have proper authorizations. Moreover, in some application scenarios, content providers may not know the exact locations of the target routers. Content providers need a way to send a message to install or remove an application module towards a network destination, and have the NetServ-enabled routers located in the path catch and act on the message. To address this challenge, we adopted on-path signaling as the deployment mechanism for NetServ. A NetServ signaling message is sent in an IP packet towards a destination. The packet gets forwarded by IP routers as usual, but when it transits a NetServ-enabled router, the message gets intercepted and passed to the NetServ control layer. Fourth, a NetServ-enabled router must support the concurrent executions of multiple content providers' applications. Each content provider's execution environment must be isolated from one another, and the resource usage of each must be controlled. To address the challenge of providing a robust multi-user execution environment, we chose to run NetServ modules in user space. This is in stark contrast to most programmable routers, which run service modules in kernel space for fast packet processing. Furthermore, NetServ modules are written in Java and run in Java Virtual Machines (JVMs). Our choice of user space execution and JVM allows us to leverage the decades of technology advances in operating systems, virtualization, and Java. Lastly, in order to host the services of a large number of content providers, NetServ must be able to scale beyond the single-box architecture. We address this challenge with the multi-box lateral expansion of NetServ using the OpenFlow forwarding engine. In this extended architecture, multiple NetServ nodes are attached to an OpenFlow switch, which provides a physically separate forwarding plane. The scalability of user services is no longer limited to a single NetServ box. Additionally, this thesis presents our prior work on improving service discovery in local and global networks. The service discovery work makes indirect contribution because the limitations of local and overlay networks encountered during those studies eventually led us to investigate in-network services, which resulted in NetServ. Specifically, we investigate the issues involved in bootstrapping large-scale structured overlay networks, present a tool to merge service announcements from multiple local networks, and propose an enhancement to structured overlay networks using link-local multicast.
     1632</li>
     1633<br>
     1634
     1635
     1636
     1637<li>
     1638<b>Lee, Jae W. and Francescangeli, Roberto and Janak, Jan and Srinivasan, Suman and Baset, Salman A. and Schulzrinne, Henning and Despotovic, Zoran and Kellerer, Wolfgang</b>
     1639, &quot;NetServ: Active Networking 2.0.&quot;
     16402011 IEEE International Conference on Communications Workshops (ICC), Kyoto, Japan, IEEE,
     16412011.
     1642doi:10.1109/iccw.2011.5963554.
     1643<a href="http://dx.doi.org/10.1109/iccw.2011.5963554">http://dx.doi.org/10.1109/iccw.2011.5963554</a>
     1644<br><br><b>Abstract: </b>We present NetServ, a node architecture for deploying in-network services in the next generation Internet. NetServ-enabled network nodes provide a common execution environment, where network services implemented as modules can be dynamically installed and removed. We demonstrate three such modules. MicroCDN is a dynamic content distribution network (CDN) service which implements a content caching strategy specific to a content provider. The NAT Keep-alive module offloads the processing of keep-alive messages from SIP servers. The Media Relay module allows any NetServ node to act as a media relay, eliminating the need to manage standalone relay servers. NetServ aims to revive the Active Networking vision. It was too far ahead of its time a decade ago, but we believe its time has finally arrived.
     1645</li>
     1646<br>
     1647
     1648
     1649
     1650<li>
     1651<b>Lee, Ki S. and Wang, Han and Weatherspoon, Hakim</b>
     1652, &quot;SoNIC: Precise Realtime Software Access and Control of Wired Networks.&quot;
     1653Proceedings of the 10th USENIX Conference on Networked Systems Design and Implementation, Lombard, IL, USENIX Association, Berkeley, CA, USA,
     16542013.
     1655
     1656<a href="http://portal.acm.org/citation.cfm?id=2482626.2482648">http://portal.acm.org/citation.cfm?id=2482626.2482648</a>
     1657<br><br><b>Abstract: </b>The physical and data link layers of the network stack contain valuable information. Unfortunately, a systems programmer would never know. These two layers are often inaccessible in software and much of their potential goes untapped. In this paper we introduce SoNIC, Software-defined Network Interface Card, which provides access to the physical and data link layers in software by implementing them in software. In other words, by implementing the creation of the physical layer bitstream in software and the transmission of this bitstream in hardware, SoNIC provides complete control over the entire network stack in realtime. SoNIC utilizes commodity off-the-shelf multi-core processors to implement parts of the physical layer in software, and employs an FPGA board to transmit optical signal over the wire. Our evaluations demonstrate that SoNIC can communicate with other network components while providing realtime access to the entire network stack in software. As an example of SoNIC's fine-granularity control, it can perform precise network measurements, accurately characterizing network components such as routers, switches, and network interface cards. Further, SoNIC enables timing channels with nanosecond modulations that are undetectable in software.
     1658</li>
     1659<br>
     1660
     1661
     1662
     1663<li>
     1664<b>Li, Dawei and Hong, Xiaoyan</b>
     1665, &quot;Practical exploitation on system vulnerability of ProtoGENI.&quot;
     1666Proceedings of the 49th Annual Southeast Regional Conference, Kennesaw, Georgia, ACM, New York, NY, USA,
     16672011.
     1668doi:10.1145/2016039.2016073.
     1669<a href="http://dx.doi.org/10.1145/2016039.2016073">http://dx.doi.org/10.1145/2016039.2016073</a>
     1670<br><br><b>Abstract: </b>Global Environment for Network Innovations (GENI) is a unique virtual laboratory for at-scale networking experimentation exploring future Internets. The successful development of GENI has to consider security problems from the design and prototyping stages. However, in many cases, system vulnerability cannot be found unless through real experimentation bearing purposeful and meaningful designs. In this paper, we introduce some of our efforts in exploring the security vulnerabilities in ProtoGENI, a prototype implementation and deployment of GENI. Our results show potential breach on security of GENI in terms of availability. We make suggestions on potential defense strategies in order to improve the ProtoGENI security and its development.
     1671</li>
     1672<br>
     1673
     1674
     1675
     1676<li>
     1677<b>Li, Dawei and Hong, Xiaoyan and Bowman, Jason</b>
     1678, &quot;Evaluation of Security Vulnerabilities by Using ProtoGENI as a Launchpad.&quot;
     1679IEEE Global Communications Conference (GLOBECOM 2011),
     16802011.
     1681
     1682<a href="ftp://202.38.75.7/pub/&#x0025;D0&#x0025;C2&#x0025;CE&#x0025;C4&#x0025;BC&#x0025;FE&#x0025;BC&#x0025;D0&#x0025;20(2)/DATA/PID1102190.PDF">ftp://202.38.75.7/pub/&#x0025;D0&#x0025;C2&#x0025;CE&#x0025;C4&#x0025;BC&#x0025;FE&#x0025;BC&#x0025;D0&#x0025;20(2)/DATA/PID1102190.PDF</a>
     1683<br><br><b>Abstract: </b>In this paper we analyze the security architecture of ProtoGENI. ProtoGENI is a prototype control framework implementation of GENI (Global Environment for Network Innovations). We perform a variety of experiments in an effort to identify potential vulnerabilities presented in the current implementation. We classify our attacks into three types: data plane to data plane, data plane to control plane, and data plane to Internet. Our results indicate the potential for a breach of confidentiality and availability internally within ProtoGENI, as well as risks to external Internet. We make suggestions outlining possible defense strategies to improve ProtoGENI security and aid in future development
     1684</li>
     1685<br>
     1686
     1687
     1688
     1689<li>
     1690<b>Li, Ting and Van Vorst, Nathanael and Liu, Jason</b>
     1691, &quot;A Rate-based TCP Traffic Model to Accelerate Network Simulation.&quot;
     1692Simulation, Society for Computer Simulation International, San Diego, CA, USA,
     16932013.
     1694doi:10.1177/0037549712469892.
     1695<a href="http://dx.doi.org/10.1177/0037549712469892">http://dx.doi.org/10.1177/0037549712469892</a>
     1696<br><br><b>Abstract: </b>Traditional discrete-event simulation of large-scale networks at the packet level is computationally expensive. This article presents a fast rate-based transmission control protocol (RTCP) traffic model designed to reduce the time and space complexity for simulating network traffic whilst maintaining good accuracy. A distinct feature of the proposed model is that the transmission control protocol (TCP) congestion control behavior is represented using analytical models that describe the send rate at the traffic source as a function of the round-trip time and the packet loss rate at different phases of a TCP connection. Rather than modeling at the granularity of individual packets visiting the intermediate routers, the model approximates traffic flows as a series of rate windows, each consisting of a number of packets considered to possess the same arrival rate. The model calculates the queuing delays and the packet losses as these rate windows traverse the individual network queues along the flow path. The proposed RTCP model is able to achieve a performance advantage over other TCP models, by integrating analytical solutions and aggregating traffic using rate windows. Empirical results show that the RTCP model can correctly capture the overall TCP behavior and achieve a speedup of more than two orders of magnitude over the corresponding detailed packet-oriented simulation.
     1697</li>
     1698<br>
     1699
     1700
     1701
     1702<li>
     1703<b>Li, Ting and Van Vorst, Nathanael and Rong, Rong and Liu, Jason</b>
     1704, &quot;Simulation studies of OpenFlow-based in-network caching strategies.&quot;
     1705Proceedings of the 15th Communications and Networking Simulation Symposium, Orlando, Florida, Society for Computer Simulation International, San Diego, CA, USA,
     17062012.
     1707
     1708<a href="http://portal.acm.org/citation.cfm?id=2331762.2331774">http://portal.acm.org/citation.cfm?id=2331762.2331774</a>
     1709<br><br><b>Abstract: </b>We propose an in-network caching architecture using Open-Flow to coordinate caching decisions in the network. Our scheme, called CacheFlow, extends the cache-and-forward concept by moving contents closer to the clients hop-by-hop using TCP for sending requests and retrieving contents. As such, CacheFlow can be incrementally implemented and deployed in the real network. In this paper, we present a simulation study of several caching policies, including a random cache policy, a statically optimal cache placement policy and a new disk placement strategy that places popular contents at the &#x63;&#x0308;enter&#x20;&#x0308;of the network. Experimental results show that simple in-network caching policies can be realized using today's technology to improve network performance.
     1710</li>
     1711<br>
     1712
     1713
     1714
     1715<li>
     1716<b>Liu, J. and Abu Obaida, M. and Dos Santos, F.</b>
     1717, &quot;Toward PrimoGENI Constellation for Distributed At-Scale Hybrid Network Test.&quot;
     1718Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     17192014.
     1720doi:10.1109/gree.2014.10.
     1721<a href="http://dx.doi.org/10.1109/gree.2014.10">http://dx.doi.org/10.1109/gree.2014.10</a>
     1722<br><br><b>Abstract: </b>PrimoGENI provides a GENI aggregate interface through which experimenters can launch large-scale network experiments on GENI resources consisting of both simulated network and real instances of network applications directly running on either virtual or physical machines. Real network traffic generated by the network applications can be introduced into the simulated network in real time and be subjected to proper delays and losses according to the simulated network conditions. To leverage the previous PrimoGENI prototype activities, PrimoGENI Constellation is a newly launched project, which will focus specifically on facilitating distributed at-scale hybrid experiments for real-world high-impact applications. In this paper, we provide an overview of the major achievements of PrimoGENI, and more importantly, discuss the ongoing efforts in PrimoGENI Constellation aiming to achieve the full potential of the hybrid network experiment approach. The main thrusts of PrimoGENI Constellation include: 1) supporting at-scale network experiments potentially distributed on different types of GENI resources in accordance with the GENI experiment workflow, 2) focusing on target applications supporting prominent and high-impact future Internet research, and 3) building the user community through extensive education and research training, and online archives of experiment results and user experiences.
     1723</li>
     1724<br>
     1725
     1726
     1727
     1728<li>
     1729<b>Liu, Jun and O'Neil, Thomas and Desell, Travis and Carlson, Ross</b>
     1730, &quot;Work-in-Progress: Empirical Verification of A Subset Sum Hypothesis in GENI Cloud.&quot;
     1731First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     17322012.
     1733
     1734
     1735
     1736</li>
     1737<br>
     1738
     1739
     1740
     1741<li>
     1742<b>Liu, Lei and Peng, Wei-Ren and Casellas, Ramon and Tsuritani, Takehiro and Morita, Itsuro and Martinez, Ricardo and Munoz, Raul and Suzuki, Masatoshi and Ben Yoo, S. J.</b>
     1743, &quot;Dynamic OpenFlow-Based Lightpath Restoration in Elastic Optical Networks on the GENI Testbed.&quot;
     1744Lightwave Technology, Journal of, IEEE,
     17452015.
     1746doi:10.1109/jlt.2014.2388194.
     1747<a href="http://dx.doi.org/10.1109/jlt.2014.2388194">http://dx.doi.org/10.1109/jlt.2014.2388194</a>
     1748<br><br><b>Abstract: </b>Elastic optical networking (EON), with its flexible use of the optical spectrum, is a promising solution for future metro/core optical networking. For the deployment of EON in a real-operational scenario, the dynamic lightpath restoration, driven by an intelligent control plane, is a necessary network function. Dynamic restoration can restore network services automatically and, thus, greatly reduce the operational cost, compared with traditional manual or semistatic lightpath restoration strategies enabled by network operators via a network management system. To this end, in this paper, we present an OpenFlow-enabled dynamic lightpath restoration in elastic optical networks, detailing the restoration framework and algorithm, the failure isolation mechanism, and the proposed OpenFlow protocol extensions. We quantitatively present the restoration performance via control plane experimental tests on the Global Environment for Network Innovations testbed.
     1749</li>
     1750<br>
     1751
     1752
     1753
     1754<li>
     1755<b>Liu, Lei and Zhu, Zuqing and Wang, Xiong and Song, Guanghua and Chen, Cen and Chen, Xiaoliang and Ma, Shoujiang and Feng, Xiaotao and Proietti, Roberto and Yoo, S. J. B.</b>
     1756, &quot;Field Trial of Broker-based Multi-domain Software-Defined Heterogeneous Wireline-Wireless-Optical Networks.&quot;
     1757Optical Fiber Communication Conference, Los Angeles, California, OSA,
     17582015.
     1759doi:10.1364/ofc.2015.th3j.5.
     1760<a href="http://dx.doi.org/10.1364/ofc.2015.th3j.5">http://dx.doi.org/10.1364/ofc.2015.th3j.5</a>
     1761<br><br><b>Abstract: </b>Driven by a broker-based OpenFlow control plane, we report the first field trial of software-defined heterogeneous wireline-wireless-optical multi-domain networks connecting UC Davis Campus, USTC, California OpenFlow Testbed Network (COTN) and Energy Sciences Network (ESNet).
     1762</li>
     1763<br>
     1764
     1765
     1766
     1767<li>
     1768<b>Liu, Xuan</b>
     1769, &quot;Dynamic Virtual Network Restoration with Optimal Standby Virtual Router Selection (Doctoral dissertation).&quot;
     1770
     17712015.
     1772
     1773<a href="https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/46697/LiuDynVirNet.pdf?sequence=1&#x0026;&#x0023;38;isAllowed=y">https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/46697/LiuDynVirNet.pdf?sequence=1&#x0026;&#x0023;38;isAllowed=y</a>
     1774<br><br><b>Abstract: </b>Network virtualization technologies allow service providers to request partitioned, QoS guaranteed and fault-tolerant virtual networks provisioned by the substrate network provider (i.e., physical infrastructure provider). A virtualized networking environment (VNE) has common features such as partition, flexibility, etc., but fault-tolerance requires additional efforts to provide survivability against failures on either virtual networks or the substrate network. Two common survivability paradigms are protection (proactive) and restoration (reactive). In the protection scheme, the substrate network provider (SNP) allocates redundant resources (e.g., nodes, paths, bandwidths, etc) to protect against potential failures in the VNE. In the restoration scheme, the SNP dynamically allocates resources to restore the networks, and it usually occurs after the failure is detected. In this dissertation, we design a restoration scheme that can be dynamically implemented in a centralized manner by an SNP to achieve survivability against node failures in the VNE. The proposed restoration scheme is designed to be integrated with a protection scheme, where the SNP allocates spare virtual routers (VRs) as standbys for the virtual networks (VN) and they are ready to serve in the restoration scheme after a node failure has been identified. These standby virtual routers (S-VR) are reserved as a sharedbackup for any single node failure, and during the restoration procedure, one of the S-VR will be selected to replace the failed VR. In this work, we present an optimal S-VR selection approach to simultaneously restore multiple VNs affected by failed VRs, where these VRs may be affected by failures within themselves or at their substrate host (i.e., power outage, hardware failures, maintenance, etc.). Furthermore, the restoration scheme is embedded into a dynamic reconfiguration scheme (DRS), so that the affected VNs can be dynamically restored by a centralized virtual network manager (VNM). We first introduce a dynamic reconfiguration scheme (DRS) against node failures in a VNE, and then present an experimental study by implementing this DRS over a realistic VNE using GpENI testbed. For this experimental study, we ran the DRS to restore one VN with a single-VR failure, and the results showed that with a proper S-VR selection, the performance of the affected VN could be well restored. Next, we proposed an Mixed-Integer Linear Programming (MILP) model with dual–goals to optimally select S-VRs to restore all VNs affected by VR failures while load balancing. We also present a heuristic algorithm based on the model. By considering a number of factors, we present numerical studies to show how the optimal selection is affected. The results show that the proposed heuristic's performance is close to the optimization model when there were sufficient standby virtual routers for each virtual network and the substrate nodes have the capability to support multiple standby virtual routers to be in service simultaneously. Finally, we present the design of a software-defined resilient VNE with the optimal S-VR selection model, and discuss a prototype implementation on the GENI testbed.
     1775</li>
     1776<br>
     1777
     1778
     1779
     1780<li>
     1781<b>Liu, Xuan and Edwards, Sarah and Riga, Niky and Medhi, Deep</b>
     1782, &quot;Design of a software-defined resilient virtualized networking environment.&quot;
     1783Design of Reliable Communication Networks (DRCN), 2015 11th International Conference on the, IEEE,
     17842015.
     1785doi:10.1109/drcn.2015.7148999.
     1786<a href="http://dx.doi.org/10.1109/drcn.2015.7148999">http://dx.doi.org/10.1109/drcn.2015.7148999</a>
     1787<br><br><b>Abstract: </b>Network virtualization enables programmability to the substrate network provider who provisions and manages virtual networks (VNs) for service providers. A mix of software-defined and autonomic technology improves the flexibility of network management, including dynamic reconfiguration in the virtualized networking environment (VNE). Virtual router (VR)s run at a logical level where software failures may be more frequent. Thus, a VR failure is more frequent than a physical router failure on the substrate network. In this paper, we present a software-defined resilient virtualized networking environment where a VN topology can be restored by using a preserved standby virtual router (S-VR) after a VR failure. We illustrate a preliminary autonomic setup of a VNE on the GENI testbed.
     1788</li>
     1789<br>
     1790
     1791
     1792
     1793<li>
     1794<b>Luna, Nicholas and Shetty, Sachin and Rogers, Tamara and Xiong, Kaiqi</b>
     1795, &quot;Assessment of Router Vulnerabilities on PlanetLab Infrastructure for Secure Cloud Computing.&quot;
     1796First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     17972012.
     1798
     1799
     1800<br><br><b>Abstract: </b>In recent times, the cloud computing based delivery model has been proven to reduce enterprise IT costs and complexities. In contrast to traditional enterprise IT solutions, the cloud computing model moves the application software and data to remote servers in large datacenters, which raises many security challenges. One of the critical challenges is the inability to characterize the impact of the vulnerabilities of routers on the cloud security and performance guarantees. In this paper, we analyze the degree of security provided by routers to data sharing applications deployed in cloud environments that span administrative and network domains. Our analysis is based on examining the security level of network applications on routers which lie between nodes on Planetlab infrastructure. We assume that some of the PlanetLab nodes will share the same wide area network path as the cloud servers. Our preliminary results confirm that the majority of the routers are plagued by insecure network protocols, leading to vulnerable routers. These results confirm our hypothesis that the security of the network infrastructure needs to be upgraded to assure the protection of information exchanged on the wide area network path.
     1801</li>
     1802<br>
     1803
     1804
     1805
     1806<li>
     1807<b>Maccherani, E. and Femminella, M. and Lee, J. W. and Francescangeli, R. and Janak, J. and Reali, G. and Schulzrinne, H.</b>
     1808, &quot;Extending the NetServ autonomic management capabilities using OpenFlow.&quot;
     18092012 IEEE Network Operations and Management Symposium, Maui, HI, IEEE,
     18102012.
     1811doi:10.1109/NOMS.2012.6211961.
     1812<a href="http://dx.doi.org/10.1109/NOMS.2012.6211961">http://dx.doi.org/10.1109/NOMS.2012.6211961</a>
     1813<br><br><b>Abstract: </b>Autonomic management capabilities of the Future Internet can be provided through a recently proposed service architecture called NetServ. It consists of the interconnection of programmable nodes which enable dynamic deployment and execution of network and application services. This paper shows how this architecture can be further improved by introducing the OpenFlow architecture and implementing the OpenFlow controller as a NetServ service, thus improving both the NetServ management performance and its flexibility. These achievements are demonstrated experimentally on the GENI environment, showing the platform self-protecting capabilities in case of a SIP DoS attack.
     1814</li>
     1815<br>
     1816
     1817
     1818
     1819<li>
     1820<b>Mahindra, R. and Bhanage, G. D. and Hadjichristofi, G. and Seskar, I. and Raychaudhuri, D. and Zhang, Y. Y.</b>
     1821, &quot;Space Versus Time Separation for Wireless Virtualization on an Indoor Grid.&quot;
     1822Next Generation Internet Networks, 2008. NGI 2008, IEEE,
     18232008.
     1824doi:10.1109/NGI.2008.36.
     1825<a href="http://dx.doi.org/10.1109/NGI.2008.36">http://dx.doi.org/10.1109/NGI.2008.36</a>
     1826<br><br><b>Abstract: </b>The decreasing cost of wireless hardware and ever increasing number of wireless testbeds has led to a shift in the protocol evaluation paradigm from simulations towards emulation. In addition, with a large number of users demanding experimental resources and lack of space and time for deploying more hardware, fair resource sharing among independent co-existing experiments is important. We study the proposed approaches to wireless virtualization with a focus on schemes conserving wireless channels rather than nodes. Our detailed comparison reveals that while experiments sharing a channel by space separation achieve better efficiency than those relying on time separation of a channel, the isolation between experiments in both cases is comparable. We propose and implement a policy manager to alleviate the isolation problem and suggest scenarios in which either of the schemes would provide a suitable virtualization solution.
     1827</li>
     1828<br>
     1829
     1830
     1831
     1832<li>
     1833<b>Mahindra, R. and Bhanage, G. and Hadjichristofi, G. and Ganu, S. and Kamat, P. and Seskar, I. and Raychaudhuri, D.</b>
     1834, &quot;Integration of heterogeneous networking testbeds.&quot;
     1835Proceedings of the 4th International Conference on Testbeds and research infrastructures for the development of networks &#x0026; communities, Innsbruck, Austria, ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), ICST, Brussels, Belgium, Belgium,
     18362008.
     1837
     1838<a href="http://portal.acm.org/citation.cfm?id=1390609">http://portal.acm.org/citation.cfm?id=1390609</a>
     1839<br><br><b>Abstract: </b>As networking research expands into new frontiers, the research community has felt a need for a heterogeneous networking research infrastructure to experiment with the interaction and integration of different types of networks, and to test the performance of various networking protocols in realistic environments. This requirement has led to the Global Environment for Network Innovations (GENI) initiative to create a global infrastructure for conducting networking experiments across diverse substrates such as wired (local and wide-area), wireless, sensor and cellular networks. In this paper, we discuss and present two models for building such an experimental infrastructure. The first model enables a wired testbed to link with wireless edge nodes during an experiment, whereas the second model enables a wireless testbed to link to wired testbeds. Proof-of-concept experiments are also presented reinforcing the usefulness of the models in terms of facilitating experiments over the integrated heterogeneous infrastructure.
     1840</li>
     1841<br>
     1842
     1843
     1844
     1845<li>
     1846<b>Malishevskiy, A. and Gurkan, D. and Dane, L. and Narisetty, R. and Narayan, S. and Bailey, S.</b>
     1847, &quot;OpenFlow-Based Network Management with Visualization of Managed Elements.&quot;
     1848Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     18492014.
     1850doi:10.1109/gree.2014.21.
     1851<a href="http://dx.doi.org/10.1109/gree.2014.21">http://dx.doi.org/10.1109/gree.2014.21</a>
     1852<br><br><b>Abstract: </b>The new software defined networking (SDN) paradigm advocates separating the data plane and the control plane, making network switches simple packet forwarding devices and leaving a logically-centralized software to control the behavior of the network. SDN introduces new possibilities for a centralized network management and configuration. The main benefit is having the programmability of the forwarding tables according to the needs of the applications. Therefore, efficient and effective management of network resources becomes even more crucial in providing effective control plane functionality to the applications. OpenFlow standardization efforts at the Open Networking Foundation resulted in an OpenFlow Configuration (OFConfig) specification to address the management of resources in networks with OpenFlow-enabled switches. We report the implementation of an intuitively easy to use interface for the OpenFlow-capable logical devices as managed resources in a SDN.
     1853</li>
     1854<br>
     1855
     1856
     1857
     1858<li>
     1859<b>Mambretti, J. and Chen, J. and Yeh, F.</b>
     1860, &quot;Software-Defined Network Exchanges (SDXs) and Infrastructure (SDI): Emerging innovations in SDN and SDI interdomain multi-layer services and capabilities.&quot;
     1861Science and Technology Conference (Modern Networking Technologies) (MoNeTeC), 2014 First International, IEEE,
     18622014.
     1863doi:10.1109/monetec.2014.6995590.
     1864<a href="http://dx.doi.org/10.1109/monetec.2014.6995590">http://dx.doi.org/10.1109/monetec.2014.6995590</a>
     1865<br><br><b>Abstract: </b>Software-Defined-Networking (SDN) is quickly transforming the networking landscape. Programmable networking based on many types of virtualization techniques, including SDN, enable high levels of abstraction for network services, control and management functions, and underlying technology resources. These approaches enable network designers to create a much wider range of services and capability, including through Software Defined Networking Exchanges (SDXs) than can be provided with traditional networks and exchange facilities, enabling a) many more dynamic provisioning options, including in real time b) faster implementation of new and enhanced services c) enabling applications, edge processes and even individuals to directly control core resources; e) substantially improved options for creating customizable networks and e) enhanced operational efficiency and effectiveness. In addition, these capabilities are now being extended to other types of Software Defined Infrastructure (SDI), including clouds, compute grids, storage devices, instruments, and many other types of edge devices.
     1866</li>
     1867<br>
     1868
     1869
     1870
     1871<li>
     1872<b>Mambretti, Joe and Chen, Jim and Yeh, Fei</b>
     1873, &quot;Software-Defined Network Exchanges (SDXs): Architecture, services, capabilities, and foundation technologies.&quot;
     1874Teletraffic Congress (ITC), 2014 26th International, IEEE,
     18752014.
     1876doi:10.1109/itc.2014.6932970.
     1877<a href="http://dx.doi.org/10.1109/itc.2014.6932970">http://dx.doi.org/10.1109/itc.2014.6932970</a>
     1878<br><br><b>Abstract: </b>Software Defined Networks (SDNs), primarily based on OpenFlow, are being deployed in single domain networks around the world. The popularity of SDNs has given rise to multiple considerations about designing, implementing, and operating Software-Defined Network Exchanges (SDXs), to enable SDNs to interconnect SDN islands and to extend SDNs across multiple domains. These goals can be accomplished only by developing new techniques that extend the single domain orientation of current SDN/OpenFlow approaches to include capabilities for multidomain control, including those for resource discovery, signaling, and dynamic provisioning. Several networking research communities have begun to investigate these concepts. Early architectural models of SDXs have been designed and implemented as prototypes. These SDXs are being used to conduct experiments and to demonstrate the potentials of SDXs.
     1879</li>
     1880<br>
     1881
     1882<li>
     1883<b>Mambretti, Joe and Chen, Jim and Yeh, Fei</b>
     1884, &quot;Creating environments for innovation: Designing and implementing advanced experimental network research testbeds based on the Global Lambda Integrated Facility and the StarLight Exchange.&quot;
     1885Computer Networks,
     18862014.
     1887doi:10.1016/j.bjp.2013.12.024.
     1888<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.024">http://dx.doi.org/10.1016/j.bjp.2013.12.024</a>
     1889<br><br><b>Abstract: </b>Large scale national and international experimental research environments are required to advance communication services and supporting network architecture, technology, and infrastructure. Theories and concepts are often explored using simulation and modeling techniques within labs or on small scale testbeds. However, while such testbeds are valuable resources for the research process, these facilities alone cannot provide an appropriate approximation of the real world conditions required to explore ideas at scale. Very large scale global, experimental network research capabilities are required to deeply investigate innovative concepts. For many years, network testbeds were created to address fairly specific, well defined, limited research goals, and they were implemented for fairly short periods. Recently, taking advantage of a number of macro information technology trends, such as virtualization and programmable resources, several network research communities have been developing innovative types of network research environments. Instead of designing traditional network testbeds, research communities are designing large scale, highly flexible distributed platforms that can be used to create many different types of testbeds. Also, rather than creating short term testbeds for limited research objectives, these new environments are being designed as long term persistent resources to support many types of experimental research. This paper describes the motivations for this trend, provides several examples of large scale distributed network research environments based on the Global Lambda Integrated Facility (GLIF) and the StarLight Exchange Facility, including the Global Environment for Network Innovation (GENI), and indicates emerging future trends for these types of environments.
     1890</li>
     1891<br>
     1892
     1893
     1894
     1895<li>
     1896<b>Mandal, A. and Ruth, P. and Baldin, I. and Xin, Yufeng and Castillo, C. and Rynge, M. and Deelman, E.</b>
     1897, &quot;Leveraging and Adapting ExoGENI Infrastructure for Data-Driven Domain Science Workflows.&quot;
     1898Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     18992014.
     1900doi:10.1109/gree.2014.12.
     1901<a href="http://dx.doi.org/10.1109/gree.2014.12">http://dx.doi.org/10.1109/gree.2014.12</a>
     1902<br><br><b>Abstract: </b>In this paper, we present our ongoing work on a novel use of networked cloud infrastructures like GENI for running adaptive domain science applications. We specifically report our recent experience at the SC'13 conference with showcasing a dynamically adaptable cloud infrastructure driven by the demand of a data-driven scientific workflow. Our work used resources from ExoGENI - a Networked Infrastructure-as-a-Service (NIaaS) testbed funded through NSF's Global Environment for Network Innovation (GENI) project. We used on-ramps to compute and data resources in the RENCI SC'13 booth to a large dynamically provisioned 'slice' spanning multiple ExoGENI cloud sites that were interconnected using dynamically provisioned connections from Internet2, NLR and ESnet. The slice was used to execute a scientific workflow driven from a computer in the RENCI SC'13 booth connected to the slice via SCinet. A closed-loop control mechanism leveraging a monitoring infrastructure based on persistent queries adapted the slice to the demands of the workflow as it executed.
     1903</li>
     1904<br>
     1905
     1906
     1907
     1908<li>
     1909<b>Mandal, A. and Xin, Yufeng and Baldine, I. and Ruth, P. and Heerman, C. and Chase, J. and Orlikowski, V. and Yumerefendi, A.</b>
     1910, &quot;Provisioning and Evaluating Multi-domain Networked Clouds for Hadoop-based Applications.&quot;
     1911Cloud Computing Technology and Science (CloudCom), 2011 IEEE Third International Conference on,
     19122011.
     1913doi:10.1109/CloudCom.2011.107.
     1914<a href="http://dx.doi.org/10.1109/CloudCom.2011.107">http://dx.doi.org/10.1109/CloudCom.2011.107</a>
     1915
     1916</li>
     1917<br>
     1918
     1919
     1920
     1921<li>
     1922<b>Mandal, Anirban and Ruth, Paul and Baldin, Ilya and Xin, Yufeng and Castillo, Claris and Rynge, Mats and Deelman, Ewa</b>
     1923, &quot;Evaluating I/O Aware Network Management for Scientific Workflows on Networked Clouds.&quot;
     1924Proceedings of the Third International Workshop on Network-Aware Data Management, Denver, Colorado, ACM, New York, NY, USA,
     19252013.
     1926doi:10.1145/2534695.2534698.
     1927<a href="http://dx.doi.org/10.1145/2534695.2534698">http://dx.doi.org/10.1145/2534695.2534698</a>
     1928<br><br><b>Abstract: </b>This paper presents a performance evaluation of scientific workflows on networked cloud systems with particular emphasis on evaluating the effect of provisioned network bandwidth on application I/O performance. The experiments were run on ExoGENI, a widely distributed networked infrastructure as a service (NIaaS) testbed. ExoGENI orchestrates a federation of independent cloud sites located around the world along with backbone circuit providers. The evaluation used a representative data-intensive scientific workflow application called Montage. The application was deployed on a virtualized HTCondor environment provisioned dynamically from the ExoGENI networked cloud testbed, and managed by the Pegasus workflow manager. The results of our experiments show the effect of modifying provisioned network bandwidth on disk I/O throughput and workflow execution time. The marginal benefit as perceived by the workflow reduces as the network bandwidth allocation increases to a point where disk I/O saturates. There is little or no benefit from increasing network bandwidth beyond this inflection point. The results also underline the importance of network and I/O performance isolation for predictable application performance, and are applicable for general data-intensive workloads. Insights from this work will also be useful for real-time monitoring, application steering and infrastructure planning for data-intensive workloads on networked cloud platforms.
     1929</li>
     1930<br>
     1931
     1932
     1933
     1934<li>
     1935<b>Mandvekar, L. and Qiao, Chunming and Husain, M. I.</b>
     1936, &quot;Enabling Wide Area Single System Image Experimentation on the GENI Platform.&quot;
     1937Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     19382013.
     1939doi:10.1109/gree.2013.27.
     1940<a href="http://dx.doi.org/10.1109/gree.2013.27">http://dx.doi.org/10.1109/gree.2013.27</a>
     1941<br><br><b>Abstract: </b>The Single System Image (SSI) clustering technology hides the distributed nature of the participating resources, and makes them appear as a single homogeneous computing resource to the user. An SSI cluster can utilize all the available processing power and memory from its participating resources. However, using the current implementations, an SSI can only be formed using nodes which are within one-hop distance of each other. This implies that nodes have to be within the same broadcast domain / Local Area Network (LAN) in order to participate in an SSI cluster. This limits the full potential of SSIs. In this research, we propose enhancements to the existing SSI technology to overcome the one-hop limitation, thus enabling nodes over a WAN to form SSI clusters. GENI provides a perfect platform for such experimentation. In this paper, we report our initial success in enabling Transparent Interprocess Communication Protocol (TIPC) over wide area nodes in GENI and progress in backporting the TIPC 2.0 protocol, with support for communication over WANs, to Kerrighed, an open-source software for creating SSIs.
     1942</li>
     1943<br>
     1944
     1945
     1946
     1947<li>
     1948<b>Mandvekar, Lokesh and Sathyaraja, Anandatirtha and Qiao, Chunming</b>
     1949, &quot;Socially Aware Single System Images.&quot;
     1950First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     19512012.
     1952
     1953
     1954<br><br><b>Abstract: </b>Cloud computing enables users to get access to huge amounts of computing resources as desired. There are many popular commercial cloud service providers which provide resources to users at a price. These providers can not be trusted as far as privacy of data is concerned. On the other hand, people do trust their close friends, relatives and other social contacts, albeit, to varying degrees. This paper reports the work-in-progress on S3I(Socially Aware Single System Images) which allows users to form computing clusters using resources owned by their social contacts. It tries to utilize the trust found between people in real life and translate it to provide trustworthy resource sharing between them.
     1955</li>
     1956<br>
     1957
     1958
     1959
     1960<li>
     1961<b>Marasevic, J. and Janak, J. and Schulzrinne, H. and Zussman, G.</b>
     1962, &quot;WiMAX in the Classroom: Designing a Cellular Networking Hands-On Lab.&quot;
     1963Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     19642013.
     1965doi:10.1109/gree.2013.29.
     1966<a href="http://dx.doi.org/10.1109/gree.2013.29">http://dx.doi.org/10.1109/gree.2013.29</a>
     1967<br><br><b>Abstract: </b>Wireless networking has recently gained tremendous attention in research and education. Since the concepts taught in wireless courses are difficult to acquire only through lectures, hands-on lab experience is indispensable. While Wi-Fi based networking labs have been introduced before, to the best of our knowledge, labs that use a cellular technology have not been designed yet. Therefore, we present a WiMAX hands-on lab designed for a graduate course in wireless and mobile networking. The lab is based on the mobile WiMAX hardware and software developed and deployed within the GENI WiMAX project. We provide a brief overview of the course and of the main concepts taught in the WiMAX lecture. Then, we describe in detail our WiMAX network and the structure of the lab experiment. The effectiveness in achieving the learning objectives is evaluated via the lab reports submitted by the students. Finally, we review some of the lessons we learned during design and implementation of this lab. These can provide important insights to designers of similar labs.
     1968</li>
     1969<br>
     1970
     1971
     1972
     1973<li>
     1974<b>Martin, Vincent and Coulaby, Adama and Schaff, Nathan and Tan, Chiu C. and Lin, Shan</b>
     1975, &quot;Bandwidth Prediction on a WiMAX Network.&quot;
     1976Mobile Ad Hoc and Sensor Systems (MASS), 2014 IEEE 11th International Conference on, IEEE,
     19772014.
     1978doi:10.1109/mass.2014.75.
     1979<a href="http://dx.doi.org/10.1109/mass.2014.75">http://dx.doi.org/10.1109/mass.2014.75</a>
     1980<br><br><b>Abstract: </b>The IEEE 802.16 standard (WiMAX) is an important next-generation networking technology which promises highspeed network access for both mobile and fixed users. In this paper we present a method to estimate link quality for devices connected to Temple University's WiMAX network as they traverse both the main campus and the city of Philadelphia via foot and motor vehicle. This is accomplished by first measuring receive signal strength indicator (RSSI), carrier to interference plus noise ratio (CINR), and bandwidth. After capturing these values, we then analyze the data to provide an estimation of the actual system rate. We then present an approach to predict future states of link quality both while stationary at Temple and when traversing Philadelphia via bus.
     1981</li>
     1982<br>
     1983
     1984
     1985
     1986<li>
     1987<b>Maziku, H. and Shetty, S.</b>
     1988, &quot;Network Aware VM Migration in Cloud Data Centers.&quot;
     1989Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     19902014.
     1991doi:10.1109/gree.2014.18.
     1992<a href="http://dx.doi.org/10.1109/gree.2014.18">http://dx.doi.org/10.1109/gree.2014.18</a>
     1993<br><br><b>Abstract: </b>Host virtualization allows data centers to live migrate an entire virtual Machine (VM) to support data center maintenance, disaster avoidance and workload balancing. Live VM Migration can consume nearly the entire bandwidth for memory intensive applications which impacts the performance of competing flows in the network. A network-aware VM Migration operation ensures a fair share allocation of network resources, leading to a seamless Virtual Machine mobility while minimizing degradation of network performance. Recently, VMPatrol was proposed as a network aware VM Migration model which uses a single physical machine and QoS policies to simulate and implement a cost of migration model. However, the performance evaluation of VMPatrol was conducted in an emulated environment. In this paper, we empirically evaluate the performance of VMPatrol in an experimental GENI testbed characterized by wide-area network dynamics and realistic traffic scenarios. We deploy OpenFlow end to end QoS policies to reserve minimum bandwidths required for successful VM Migration. Preliminary results demonstrate that enforcing QoS policies in terms of bandwidth reservation relieves the network of possible overloads during migration. The results indicate that time taken to complete VM Migration depends on VM's memory size, VM page dirty rate and the available bandwidth. The results also indicate that length of stop copy phase and minimum required progress amount are critical parameters in estimating the VM migration cost.
     1994</li>
     1995<br>
     1996
     1997
     1998
     1999<li>
     2000<b>Maziku, Hellen and Shetty, Sachin and Rogers, Tamara</b>
     2001, &quot;Measurement-based IP Geolocation of Routers on Planetlab Infrastructure.&quot;
     2002First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     20032012.
     2004
     2005
     2006<br><br><b>Abstract: </b>Location aware applications can benefit from a more accurate yet robust IP geolocation framework. Various approaches to IP geolocation have been well documented. The most recent approach casts IP geolocation as a machine learn- ing classification problem. This approach makes it possible to incorporate both delay and non delay based information. The accuracy of IP geolocation can be improved by incorporating additional types of geolocation information rather relying on network delay alone. To enhance the classification accuracy of the existing classification framework, we expand it to include 6 features (3 of which are novel). We use PlanetLab as a testbed to generate our measurement set. We select 67 PlanetLab nodes within the United States with known geographic location as our landmarks. We test the accuracy of our framework on 23,843 routers given ping measurements from the 67 landmarks. With only three features (average delay, average hops and population density) tested, our new classifier gives a reduced average error distance of 157.81 miles and a median error distance of 0 miles, compared to the present classifier that gives an average error distance of 253.34 miles. This is very promising as we move on to the next phase of incorporating data for the remaining 5 features. To the best of our knowledge, this is the first proposed framework that aims to improve the accuracy of the present classifier based IP geolocation.
     2007</li>
     2008<br>
     2009
     2010
     2011
     2012<li>
     2013<b>McKeown, Nick and Anderson, Tom and Balakrishnan, Hari and Parulkar, Guru and Peterson, Larry and Rexford, Jennifer and Shenker, Scott and Turner, Jonathan</b>
     2014, &quot;OpenFlow: enabling innovation in campus networks.&quot;
     2015SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
     20162008.
     2017doi:10.1145/1355734.1355746.
     2018<a href="http://doi.acm.org/10.1145/1355734.1355746">http://doi.acm.org/10.1145/1355734.1355746</a>
     2019<br><br><b>Abstract: </b>This whitepaper proposes OpenFlow: a way for researchers to run experimental protocols in the networks they use every day. OpenFlow is based on an Ethernet switch, with an internal flow-table, and a standardized interface to add and remove flow entries. Our goal is to encourage networking vendors to add OpenFlow to their switch products for deployment in college campus backbones and wiring closets. We believe that OpenFlow is a pragmatic compromise: on one hand, it allows researchers to run experiments on heterogeneous switches in a uniform way at line-rate and with high port-density; while on the other hand, vendors do not need to expose the internal workings of their switches. In addition to allowing researchers to evaluate their ideas in real-world traffic settings, OpenFlow could serve as a useful campus component in proposed large-scale testbeds like GENI. Two buildings at Stanford University will soon run OpenFlow networks, using commercial Ethernet switches and routers. We will work to encourage deployment at other schools; and We encourage you to consider deploying OpenFlow in your university network too.
     2020</li>
     2021<br>
     2022
     2023
     2024
     2025<li>
     2026<b>Medhi, Deep and Ramamurthy, Byrav and Scoglio, Caterina and Rohrer, Justin P. and &#x43;&#x0327;etinkaya, Egemen K. and Cherukuri, Ramkumar and Liu, Xuan and Angu, Pragatheeswaran and Bavier, Andy and Buffington, Cort and Sterbenz, James P. G.</b>
     2027, &quot;The GpENI testbed: Network infrastructure, implementation experience, and experimentation.&quot;
     2028Computer Networks,
     20292014.
     2030doi:10.1016/j.bjp.2013.12.027.
     2031<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.027">http://dx.doi.org/10.1016/j.bjp.2013.12.027</a>
     2032<br><br><b>Abstract: </b>The Great Plains Environment for Network Innovation (GpENI) is an international programmable network testbed centered initially in the Midwest US with the goal to provide programmability across the entire protocol stack. In this paper, we present the overall GpENI framework and our implementation experience for the programmable routing environment and the dynamic circuit network (DCN). GpENI is built to provide a collaborative research infrastructure enabling the research community to conduct experiments in Future Internet architecture. We present illustrative examples of our experimentation in the GpENI platform.
     2033</li>
     2034<br>
     2035
     2036
     2037
     2038<li>
     2039<b>Mekky, H. and Jin, Cheng and Zhang, Zhi-Li</b>
     2040, &quot;VIRO-GENI: SDN-Based Approach for a Non-IP Protocol in GENI.&quot;
     2041Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     20422014.
     2043doi:10.1109/gree.2014.14.
     2044<a href="http://dx.doi.org/10.1109/gree.2014.14">http://dx.doi.org/10.1109/gree.2014.14</a>
     2045<br><br><b>Abstract: </b>Non-IP protocols always presented a challenge for network researchers to deploy and test at large scale. GENI infrastructure presents a testbed to deploy large scale network experiments, however, non-IP protocols still raises a challenge to deploy since IP is the narrow waist of the Internet. SDN provides an opportunity implement non-IP protocols, however, the OpenFlow standard is still tied to Ethernet/IP/TCP protocol stack. In the paper, we utilize SDN to provide a framework to deploy and test a non-IP protocol, Virtual Id Routing (VIRO), in GENI using an extended Open vSwitch platform.
     2046</li>
     2047<br>
     2048
     2049
     2050
     2051<li>
     2052<b>Mitroff, Sarah</b>
     2053, &quot;Lawrence Landweber Helped Build Today's Internet, Now He's Advising Its Future.&quot;
     2054Wired,
     20552012.
     2056
     2057<a href="http://www.wired.com/business/2012/08/lawrence-landweber/">http://www.wired.com/business/2012/08/lawrence-landweber/</a>
     2058
     2059</li>
     2060<br>
     2061
     2062
     2063
     2064<li>
     2065<b>Muhammad, Monzur and Cappos, Justin</b>
     2066, &quot;Towards a Representive Testbed: Harnessing Volunteers for Networks Research.&quot;
     2067First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     20682012.
     2069
     2070
     2071<br><br><b>Abstract: </b>A steady rise in home systems has been seen over the past few years. As more systems are designed and deployed, an appropriate testbed is required to test these systems. Sev- eral systems exist, such as PlanetLab, that currently provide a networking testbed allowing researchers and developers to test and measure various applications. However in the long run such testbeds will be unable to keep up and meet all the demands of many of the large scale modern day peer-to-peer systems. We outline the various challenges and essentials of a networking testbed and we provide an alternate network- ing testbed that is driven by resources that are voluntarily contributed. We talk about the various advantages and dis- advantages of the Seattle system, an open source peer-to- peer computing testbed that has the potential to meet these demands. The testbed is composed of sandboxed resources that are donated by volunteers. Seattle has been deployed for about three years and supports many researchers who are interested in a networking testbed. The testbed consists of over 4100 nodes and is constantly growing. Seattle looks to grow and meet the demands of networking testbeds as they are made.
     2072</li>
     2073<br>
     2074
     2075
     2076
     2077<li>
     2078<b>Mukherjee, Shreyasee and Baid, Akash and Raychaudhuri, Dipankar</b>
     2079, &quot;Integrating Advanced Mobility Services into the Future Internet Architecture.&quot;
     20807th International Conference on COMmunication Systems &#x0026; NETworkS (COMSNETS 2015), Bangalore,
     20812015.
     2082
     2083<a href="http://winlab.rutgers.edu/&#x73;&#x0303;hreya/comsnets.pdf">http://winlab.rutgers.edu/&#x73;&#x0303;hreya/comsnets.pdf</a>
     2084<br><br><b>Abstract: </b>This paper discusses the design challenges associated with supporting advanced mobility services in the future Internet. The recent transition of the Internet from the fixed host-server model to one in which mobile platforms are the norm motivates a next-generation protocol architecture which provides integrated and efficient support for advanced mobility services. Key wireless access and mobility usage scenarios are identified including host mobility, multihoming, vehicular access and context addressability, and key protocol support requirements are identified in each case. The MobilityFirst (MF) architecture being developed under the National Science Foundation's future Internet Architecture (FIA) program is proposed as a possible realization that meets the identified requirements. MF protocol specifics are given for each wireless/mobile use case, along with sample evaluation results demonstrating achievable performance benefits.
     2085</li>
     2086<br>
     2087
     2088
     2089
     2090<li>
     2091<b>Narisetty, R. and Dane, L. and Malishevskiy, A. and Gurkan, D. and Bailey, S. and Narayan, S. and Mysore, S.</b>
     2092, &quot;OpenFlow Configuration Protocol: Implementation for the of Management Plane.&quot;
     2093Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     20942013.
     2095doi:10.1109/gree.2013.21.
     2096<a href="http://dx.doi.org/10.1109/gree.2013.21">http://dx.doi.org/10.1109/gree.2013.21</a>
     2097<br><br><b>Abstract: </b>Separation of data and control plane offers benefits of having programmability of the forwarding tables according to the needs of the applications. The need for efficient and effective management of network resources is crucial in providing effective control plane functionality to the applications. OpenFlow standardization efforts at Open Networking Foundation resulted in an OpenFlow Configuration specification to address the management of resources in OpenFlow-enabled switches. We report the implementation of the OF-Config 1.1 standard [revision - 25th June 2012] as softconf.d to retrieve and update the controller IP of an OpenvSwitch.
     2098</li>
     2099<br>
     2100
     2101
     2102
     2103<li>
     2104<b>Narisetty, RajaRevanth and Gurkan, Deniz</b>
     2105, &quot;Identification of network measurement challenges in OpenFlow-based service chaining.&quot;
     2106Local Computer Networks Workshops (LCN Workshops), 2014 IEEE 39th Conference on, IEEE,
     21072014.
     2108doi:10.1109/lcnw.2014.6927718.
     2109<a href="http://dx.doi.org/10.1109/lcnw.2014.6927718">http://dx.doi.org/10.1109/lcnw.2014.6927718</a>
     2110<br><br><b>Abstract: </b>Software-defined networking and Network Function Virtualization (NFV) have simplified the coordination efforts for ” service chaining.” Consequently, network services such as firewall, load balancing, etc. may be service chained in the forwarding (data) plane for specific applications and/or traffic. A specific case is for the firewall rules that depend on deep packet inspection for application identification. If a particular application is identified and is ” safe,” would it be worthwhile to program the data plane to bypass the FW for the duration of the application session? For such a traffic-steering case, we report measurement challenges on various setups and the related cost analysis based on the network delay. Measurements of the network and processing delay have been performed with virtualized resources, on GENI testbed, and with isolated hardware units. Experiences are also reported on how a commercial firewall virtual appliance has been deployed on the GENI testbed for experimentation. The results illustrate the measurement uncertainties and challenges for DPI-based traffic steering in virtualized environments. In addition, we show that such a service chaining may increase throughput and relieve DPI-based processing overhead on firewall units.
     2111</li>
     2112<br>
     2113
     2114
     2115
     2116<li>
     2117<b>Navaz, Abdul and Velusam, Gandhimathi and Gurkan, Deniz</b>
     2118, &quot;Experiments on Networking of Hadoop.&quot;
     2119Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     21202014.
     2121doi:10.1109/icnp.2014.87.
     2122<a href="http://dx.doi.org/10.1109/icnp.2014.87">http://dx.doi.org/10.1109/icnp.2014.87</a>
     2123<br><br><b>Abstract: </b>Hadoop is a popular application process big data problems in a networked dist computers. Investigations of performance for networking have been of interest with the networking paradigm through on-demand an enforcements. Network usage characterization can further help understand what policy info needed during application use cases. At scale e Hadoop jobs will help facilitate such char report how Hadoop networking usage can be chi experimentation environment using the Environment for Network Innovation). Further distributed switch framework that may help alleviate the fault tolerance schemes in Hadoop application in the forwarding plane. Delay in recovery from failures has been reduced by almost 99\\ through such a distributed switch architecture deployed on the GENT experimentation environment.
     2124</li>
     2125<br>
     2126
     2127
     2128
     2129<li>
     2130<b>Naylor, David and Mukerjee, Matthew K. and Agyapong, Patrick and Grandl, Robert and Kang, Ruogu and Machado, Michel and Brown, Stephanie and Doucette, Cody and Hsiao, Hsu C. and Han, Dongsu and Kim, Tiffany H. and Lim, Hyeontaek and Ovon, Carol and Zhou, Dong and Lee, Soo B. and Lin, Yue H. and Stuart, Colleen and Barrett, Daniel and Akella, Aditya and Andersen, David and Byers, John and Dabbish, Laura and Kaminsky, Michael and Kiesler, Sara and Peha, Jon and Perrig, Adrian and Seshan, Srinivasan and Sirbu, Marvin and Steenkiste, Peter</b>
     2131, &quot;XIA: Architecting a More Trustworthy and Evolvable Internet.&quot;
     2132SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
     21332014.
     2134doi:10.1145/2656877.2656885.
     2135<a href="http://dx.doi.org/10.1145/2656877.2656885">http://dx.doi.org/10.1145/2656877.2656885</a>
     2136<br><br><b>Abstract: </b>Motivated by limitations in today's host-centric IP network, recent studies have proposed clean-slate network architectures centered around alternate first-class principals, such as content, services, or users. However, muchlike the host-centric IP design, elevating one principal type above others hinders communication between other principals and inhibits the network's capability to evolve. This paper presents the eXpressive Internet Architecture (XIA), an architecture with native support for multiple principals and the ability to evolve its functionality to accommodate new, as yet unforeseen, principals over time. We present the results of our ongoing research motivated by and building on the XIA architecture, ranging from topics at the physical level (``how fast can XIA go'') up through to the user level.
     2137</li>
     2138<br>
     2139
     2140
     2141
     2142<li>
     2143<b>Nozaki, Yoshihiro and Bakshi, Parth and Tuncer, Hasan and Shenoy, Nirmala</b>
     2144, &quot;Evaluation of tiered routing protocol in floating cloud tiered internet architecture.&quot;
     2145Computer Networks,
     21462014.
     2147doi:10.1016/j.bjp.2013.11.010.
     2148<a href="http://dx.doi.org/10.1016/j.bjp.2013.11.010">http://dx.doi.org/10.1016/j.bjp.2013.11.010</a>
     2149<br><br><b>Abstract: </b>Clean slate future Internet initiatives have been ongoing for a few years. An important consideration in the eventual deployment of solutions for such Internet architectures is the testing and validation of the design and its scalability in realistic network environments. Large scale emulation and experimentation testbeds sponsored and funded by major research organizations worldwide provide a suitable platform for the purpose. In this article, we present the implementation details of a new network and routing protocol that entirely replaces IP and its routing protocols from the protocol stack to provide efficient routing and forwarding of packets in a clean slate Floating Cloud Tiered (FCT) Internet architecture. The FCT architecture leverages the tier structure existing among ISPs, and has a new addressing and routing schema based on tiers. In this article, the implementation and evaluation details of the network protocol with these two features, namely the tiered addressing and tier-based routing using the Global Environmental for Network Innovations (GENI) testbed are presented. The performance of the protocol is also compared with Open Shortest Path First (OSPF) implemented over the GENI testbed for identical network topologies.
     2150</li>
     2151<br>
     2152
     2153
     2154
     2155<li>
     2156<b>O'Neill, Derek and Aikat, Jay and Jeffay, Kevin</b>
     2157, &quot;Experiment Replication Using ProtoGENI nodes.&quot;
     21582013 Second GENI Research and Educational Experiment Workshop, Salt Lake, UT, USA, IEEE,
     21592013.
     2160doi:10.1109/gree.2013.11.
     2161<a href="http://dx.doi.org/10.1109/gree.2013.11">http://dx.doi.org/10.1109/gree.2013.11</a>
     2162<br><br><b>Abstract: </b>Repeatability of network experiments has long been a goal for networking researchers but the lack of a scientific process of experimentation has made this exercise difficult to achieve. In this paper, we demonstrate that, if conducted in a scientific manner, experiments can indeed be repeated in different networks to produce the same results. We ran experiments in our lab, and on two different network configurations in ProtoGENI and demonstrated that we get similar results for network and application performance evaluations. We believe that this is an important step as we take the process of measurement-based networking research from its ad hoc phase into the realm of a scientific process. We also present these experiments using GENI infrastructure as a demonstration for other experimenters to run similar realistic experiments on GENI testbeds. Furthermore, we believe this is the first set of experiments to emulate per-connection Round- Trip-Times in GENI-based experiments.
     2163</li>
     2164<br>
     2165
     2166
     2167
     2168<li>
     2169<b>Ozcelik, I. and Fu, Yu and Brooks, R. R.</b>
     2170, &quot;DoS Detection is Easier Now.&quot;
     2171Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     21722013.
     2173doi:10.1109/gree.2013.18.
     2174<a href="http://dx.doi.org/10.1109/gree.2013.18">http://dx.doi.org/10.1109/gree.2013.18</a>
     2175<br><br><b>Abstract: </b>In this study, we test anomaly based Denial of Service (DoS) detection approaches on networks with different utilization profiles. In the experiments, we use operational background traffic and performed Distributed DoS attacks without disturbing the operational network. Experiment results indicate that the detection approach's detection performance is inversely proportional to network utilization and optimal detection parameters depend on network utilization.
     2176</li>
     2177<br>
     2178
     2179
     2180
     2181<li>
     2182<b>Ozcelik, Ilker and Brooks, Richard R.</b>
     2183, &quot;Performance Analysis of DDoS Detection Methods on Real Network.&quot;
     2184First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     21852012.
     2186
     2187
     2188<br><br><b>Abstract: </b>Distributed Denial of Service (DDoS) attacks are major security threats to the Internet. The distributed structure of these attacks makes it difficult to distinguish between legitimate and attack traffic, making detection difficult. In addition to this challenge, researchers also have to study and find countermeasures against these attacks without using an operational network for testing, since attacks on operational networks inconvenience users. In this paper, we propose a method to perform DDoS analysis on real hardware using real traffic without jeopardizing the original network. We implement our experiments on the Geni testbed using Openflow. We present results from DDoS detection methods using operational traffic.
     2189</li>
     2190<br>
     2191
     2192<li>
     2193<b>Ozcelik, Ilker and Brooks, Richard R.</b>
     2194, &quot;Operational System Testing for Designed in Security.&quot;
     2195Proceedings of the Eighth Annual Cyber Security and Information Intelligence Research Workshop, Oak Ridge, Tennessee, ACM, New York, NY, USA,
     21962013.
     2197doi:10.1145/2459976.2460038.
     2198<a href="http://dx.doi.org/10.1145/2459976.2460038">http://dx.doi.org/10.1145/2459976.2460038</a>
     2199<br><br><b>Abstract: </b>To design secure systems, one needs to understand how attackers use system vulnerabilities in their favor. This requires testing vulnerabilities on operational systems. However, working on operational systems is not always possible because of the risk of disturbance. In this study, we introduce an approach to experimenting using operational system data and performing real attacks without disturbing the original system. We applied this approach to a network security experiment and tested the performance of three detection methods. The approach used in this study can be used when developing systems with Designed-in Security to identify and test system vulnerabilities.
     2200</li>
     2201<br>
     2202
     2203<li>
     2204<b>Ozcelik, Ilker and Brooks, Richard R.</b>
     2205, &quot;Security experimentation using operational systems.&quot;
     2206Proceedings of the Seventh Annual Workshop on Cyber Security and Information Intelligence Research, Oak Ridge, Tennessee, ACM, New York, NY, USA,
     22072011.
     2208doi:10.1145/2179298.2179388.
     2209<a href="http://dx.doi.org/10.1145/2179298.2179388">http://dx.doi.org/10.1145/2179298.2179388</a>
     2210<br><br><b>Abstract: </b>Computers and Internet have evolved into necessary tools for our professional, personal and social lives. As a result of this growing dependence, there is a concern that these systems remain protected and available. This concern increases exponentially when considering systems such as smart power grids. Therefore, research should be conducted to develop effective ways of detecting system anomalies. To have realistic results, the studies should be tested on real systems. However, it is not possible to test these experiments on the live network. With the recent collaboration of Universities and research labs, a new experiment test bed has been established. As a result, experiments can now be implemented on real networks. In our study, we design an experiment to analyze Distributed Denial of Service Attacks (DDoS Attack) on a real network with real Internet traffic. The approach that we use in our study can easily be generalized to apply to smart power grids.
     2211</li>
     2212<br>
     2213
     2214
     2215
     2216<li>
     2217<b>Patali, Rohit</b>
     2218, &quot;Utility-Directed Resource Allocation in Virtual Desktop Clouds (Master's thesis).&quot;
     2219
     22202011.
     2221
     2222<a href="https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1306872632">https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1306872632</a>
     2223<br><br><b>Abstract: </b>User communities are rapidly transitioning their ” traditional desktops” that have dedicated hardware and software installations into ” virtual desktop clouds” (VDCs) that are accessible via thin-clients. To allocate and manage VDC resources for Internet-scale desktop delivery, existing work focuses mainly on managing server-side resources based on utility functions of CPU and memory loads, and do not consider network health and thin-client user experience. Resource allocations without combined utility-directed information of system loads, network health and thin-client user experience in VDC platforms inevitably results in costly guesswork and overprovisioning of resources. In this thesis, an analytical model i.e., ” Utility-Directed Resource Allocation Model (U-RAM)” is presented to solve the combined utility-directed resource allocation problem within VDCs. The solution uses an iterative algorithm that leverages utility functions of system, network and human components obtained using a novel virtual desktop performance benchmarking toolkit i.e., ” VDBench”. The combined utility functions are used to direct decision schemes based on Kuhn-Tucker optimality conditions for creating user desktop pools and determining optimal resource allocation size/location. U-RAM is evaluated in a VDC testbed featuring: (a) popular user applications (Spreadsheet Calculator, Internet Browser, Media Player, Interactive Visualization), and (b) TCP/UDP based thin-client protocols (RDP, RGS, PCoIP) under a variety of user load and network health conditions. Evaluation results demonstrate that U-RAM solution maximizes VDC scalability i.e., 'VDs per core density', and 'user connections quantity', while delivering satisfactory thin-client user experience.
     2224</li>
     2225<br>
     2226
     2227
     2228
     2229<li>
     2230<b>Paul, Subharthi and Pan, Jianli and Jain, Raj</b>
     2231, &quot;Architectures for the future networks and the next generation Internet: A survey.&quot;
     2232Computer Communications, Elsevier Science Publishers B. V., Amsterdam, The Netherlands, The Netherlands,
     22332011.
     2234doi:10.1016/j.comcom.2010.08.001.
     2235<a href="http://dx.doi.org/10.1016/j.comcom.2010.08.001">http://dx.doi.org/10.1016/j.comcom.2010.08.001</a>
     2236<br><br><b>Abstract: </b>Networking research funding agencies in USA, Europe, Japan, and other countries are encouraging research on revolutionary networking architectures that may or may not be bound by the restrictions of the current TCP/IP based Internet. We present a comprehensive survey of such research projects and activities. The topics covered include various testbeds for experimentations for new architectures, new security mechanisms, content delivery mechanisms, management and control frameworks, service architectures, and routing mechanisms. Delay/disruption tolerant networks which allow communications even when complete end-to-end path is not available are also discussed.
     2237</li>
     2238<br>
     2239
     2240
     2241
     2242<li>
     2243<b>Peter, Simon and Javed, Umar and Zhang, Qiao and Woos, Doug and Anderson, Thomas and Krishnamurthy, Arvind</b>
     2244, &quot;One tunnel is (often) enough.&quot;
     2245Proceedings of the ACM SIGCOMM 2014 conference, ACM, New York, NY, USA,
     22462014.
     2247doi:10.1145/2740070.2626318.
     2248<a href="http://dx.doi.org/10.1145/2740070.2626318">http://dx.doi.org/10.1145/2740070.2626318</a>
     2249<br><br><b>Abstract: </b>A longstanding problem with the Internet is that it is vulnerable to outages, black holes, hijacking and denial of service. Although architectural solutions have been proposed to address many of these issues, they have had difficulty being adopted due to the need for widespread adoption before most users would see any benefit. This is especially relevant as the Internet is increasingly used for applications where correct and continuous operation is essential. In this paper, we study whether a simple, easy to implement model is sufficient for addressing the aforementioned Internet vulnerabilities. Our model, called ARROW (Advertised Reliable Routing Over Waypoints), is designed to allow users to configure reliable and secure end to end paths through participating providers. With ARROW, a highly reliable ISP offers tunneled transit through its network, along with packet transformation at the ingress, as a service to remote paying customers. Those customers can stitch together reliable end to end paths through a combination of participating and non-participating ISPs in order to improve the fault-tolerance, robustness, and security of mission critical transmissions. Unlike efforts to redesign the Internet from scratch, we show that ARROW can address a set of well-known Internet vulnerabilities, for most users, with the adoption of only a single transit ISP. To demonstrate ARROW, we have added it to a small-scale wide-area ISP we control. We evaluate its performance and failure recovery properties in both simulation and live settings.
     2250</li>
     2251<br>
     2252
     2253
     2254
     2255<li>
     2256<b>Qin, Z. and Xiong, X. and Chuah, M.</b>
     2257, &quot;Lehigh Explorer: Android Application Utilizing Content Centric Features.&quot;
     2258First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     22592012.
     2260
     2261
     2262<br><br><b>Abstract: </b>Companies, government organizations or institutions from anywhere in the world publish different types of information e.g. news, health alerts, disaster warnings at any time. Rather than consuming all published data, users only desire access to information of interest to themselves irrespective of where the data is located and who publish them. Existing publish/subscribe systems built based on IP-based network can be inefficient and are not flexible enough to meet emerging requirements e.g. deal with mobile users, dynamic contents, searching over encrypted data. Recently content-centric networks have been proposed to provide flexibility to users to access such information. We have designed secure content centric mobile networks that allow users to publish and retrieve contents securely. As with any new architecture, one important issue is to have useful applications that can utilize features provided in the new architecture. In this paper, we describe an Android application we recently developed that allows visitors to explore Lehigh campus based on their expressed interests. Our application utilizes keyword based interest messages to retrieve matching data items of interests to a user. We are giving a demo of Lehigh Explorer at GEC13.
     2263</li>
     2264<br>
     2265
     2266
     2267
     2268<li>
     2269<b>Qiu, Chenxi and Shen, Haiying</b>
     2270, &quot;A Delaunay-Based Coordinate-Free Mechanism for Full Coverage in Wireless Sensor Networks.&quot;
     2271Parallel and Distributed Systems, IEEE Transactions on, IEEE,
     22722014.
     2273doi:10.1109/tpds.2013.134.
     2274<a href="http://dx.doi.org/10.1109/tpds.2013.134">http://dx.doi.org/10.1109/tpds.2013.134</a>
     2275<br><br><b>Abstract: </b>Recently, many schemes have been proposed for detecting and healing coverage holes to achieve full coverage in wireless sensor networks (WSNs). However, none of these schemes aim to find the shortest node movement paths to heal the coverage holes, which could significantly reduce energy usage for node movement. Also, current hole healing schemes require accurate knowledge of sensor locations; obtaining this knowledge consumes high energy. In this paper, we propose a Delaunay-based coordinate-free mechanism (DECM) for full coverage. Based on rigorous mathematical analysis, DECM can detect coverage holes and find the locally shortest paths for healing holes in a distributed manner without requiring accurate node location information. Also, DECM incorporates a cooperative movement mechanism that can prevent generating new holes during node movements in healing holes. Simulation results and experimental results from the real-world GENI Orbit testbed show that DECM achieves superior performance in terms of the energy-efficiency, effectiveness of hole healing, energy consumption balance and lifetime compared to previous schemes.
     2276</li>
     2277<br>
     2278
     2279
     2280
     2281<li>
     2282<b>Quan, John and Nance, Kara and Hay, Brian</b>
     2283, &quot;A Mutualistic Security Service Model: Supporting Large-Scale Virtualized Environments.&quot;
     2284IT Professional,
     22852011.
     2286doi:10.1109/MITP.2011.36.
     2287<a href="http://dx.doi.org/10.1109/MITP.2011.36">http://dx.doi.org/10.1109/MITP.2011.36</a>
     2288<br><br><b>Abstract: </b>Applying a mutualistic security service model to large-scale virtualized environments that rely on contributed hardware lets researchers improve security in exchange for resources. The authors discuss this model in the context of the Global Environment for Network Innovation (GENI) project.
     2289</li>
     2290<br>
     2291
     2292
     2293
     2294<li>
     2295<b>Rajagopalan, Sudharsan</b>
     2296, &quot;Leveraging OpenFlow for Resource Placement of Virtual Desktop Cloud Applications.&quot;
     2297
     22982013.
     2299
     2300<a href="http://rave.ohiolink.edu/etdc/view?acc&#x005F;num=osu1367456412">http://rave.ohiolink.edu/etdc/view?acc&#x005F;num=osu1367456412</a>
     2301<br><br><b>Abstract: </b>Popular applications such as email, photo/video galleries, and file storage are increasingly being supported by cloud platforms in residential, academia and industry communities. The next frontier for these user communities will be to transition `traditional desktops' that have dedicated hardware and software configurations into `virtual desktop clouds' that are accessible via thin-clients. In this paper, we describe an Intelligent resource placement framework for thin-client based virtual desktops. The framework leverages principles of software defined networking and features a `unified resource broker' that uses special `marker packets' for: (a) ” route setup” when handling non-IP traffic between thin-client sites and data centers, (b) ” path selection” and ” load balancing” of virtual desktop flows to improve performance of interactive applications and video playback, and to cope with faults such as link-failures or Denial of Service cyber-attacks. The Framework has the ability to provisioning OpenFlow paths with less Service Response times for VD Requests. Our Framework In addition, we detail our framework implementation within a virtual desktop cloud (VDC) setup in a multi-domain Global Environment for Network Innovations (GENI) Future Internet testbed spanning backbone and access networks with a automation and centralized control using a tool called VDC-Sim. We present empirical results from our experimentation that leverages OpenFlow programmable networking, as well as cross-traffic capabilities for validating our framework in GENI under realistic settings. Our results demonstrate the importance of scheduling regulated measurements that can be used for intelligent resource placement decisions. Our results also show the feasibility and benefits of using OpenFlow controller applications for path selection and load balancing between thin-client sites and data centers in VDCs. The thesis also shows how our OpenFlow Framework can used for other cloud applications using GridFTP application over WAN as a Case Study.
     2302</li>
     2303<br>
     2304
     2305
     2306
     2307<li>
     2308<b>Rakotoarivelo, Thierry and Jourjon, Guillaume and Mehani, Olivier and Ott, Maximilian and Zink, Mike</b>
     2309, &quot;Repeatable Experiments with LabWiki.&quot;
     2310
     23112014.
     2312
     2313<a href="http://arxiv.org/abs/1410.1681">http://arxiv.org/abs/1410.1681</a>
     2314<br><br><b>Abstract: </b>The ability to repeat the experiments from a research study and obtain similar results is a corner stone in experiment-based scientific discovery. This essential feature has been often ignored by the distributed computing and networking community. There are many reasons for that, such as the complexity of provisioning, configuring, and orchestrating the resources used by experiments, their multiple external dependencies, and the difficulty to seamlessly record these dependencies. This paper describes a methodology based on well-established principles to plan, prepare and execute experiments. We propose and describe a family of tools, the LabWiki workspace, to support an experimenter's workflow based on that methodology. This proposed workspace provides services and mechanisms for each step of an experiment-based study, while automatically capturing the necessary information to allow others to repeat, inspect, validate and modify prior experiments. Our LabWiki workspace builds on existing contributions, and de-facto protocol and model standards, which emerged from recent experimental facility initiatives. We use a real experiment as a thread to guide and illustrate the discussion throughout this paper.
     2315</li>
     2316<br>
     2317
     2318
     2319
     2320<li>
     2321<b>Ramisetty, Shravya and Calyam, Prasad and Cecil, J. and Akula, Amit R. and Antequera, Ronny B. and Leto, Ray</b>
     2322, &quot;Ontology integration for advanced manufacturing collaboration in cloud platforms.&quot;
     2323Integrated Network Management (IM), 2015 IFIP/IEEE International Symposium on, IEEE,
     23242015.
     2325doi:10.1109/inm.2015.7140329.
     2326<a href="http://dx.doi.org/10.1109/inm.2015.7140329">http://dx.doi.org/10.1109/inm.2015.7140329</a>
     2327<br><br><b>Abstract: </b>Advances in the field of cloud computing and networking have led to rapid development and market growth in areas such as online retail, gaming and healthcare. In the field of advanced manufacturing however, the impact has been significantly lesser than expected due to limitations in cloud platforms for fostering community engagement. To address this problem, we study a new cloud-based architecture that provides Platform-asa-Service (PaaS) management capabilities to the manufacturing community for delivering Software-as-a-Service (SaaS) ” Apps” to their customers. Our architecture aims at supporting an ” App Marketplace” that thrives on agile development, organic collaboration and scalable sales of next generation manufacturing Apps requiring high-performance simulation and modeling. Towards realizing the vision of the above architecture, our paper involves investigation and implementation of an Ontology Service that interoperates with other common web services related to resource brokering and accounting. Our Ontology Service uses principles of mapping and merging to translate a manufacturing App's collaboration requirements to suitable resource specifications on public cloud platforms. Integrated resultant ontology can be queried to provision the required resource parameters such as amount of memory/storage, number of processing units, and network protocol configurations needed for deployment of an App. We validate the effectiveness of our Ontology Service using the Protégé framework in a pilot testbed of a real-world ” WheelSim” App in the NSF GENI Cloud platform. Our ontology integration results show benefits to an App developer in terms of: optimal user experience, lower design time and lower cost/simulation.
     2328</li>
     2329<br>
     2330
     2331
     2332
     2333<li>
     2334<b>Randall, David P. and Diamant, E. Ilana and Lee, Charlotte P.</b>
     2335, &quot;Creating Sustainable Cyberinfrastructures.&quot;
     2336Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, Republic of Korea, ACM, New York, NY, USA,
     23372015.
     2338doi:10.1145/2702123.2702216.
     2339<a href="http://dx.doi.org/10.1145/2702123.2702216">http://dx.doi.org/10.1145/2702123.2702216</a>
     2340<br><br><b>Abstract: </b>In this paper we report the results of a qualitative research study of the GENI cyberinfrastructure: a program of four federated cyberinfrastructures. Drawing on theories of stakeholder positioning, we examine how different GENI stakeholders attempt to enlist new participants in the cyberinfrastructures of GENI, and leverage existing relationships to create sustainable infrastructure. This study contributes to our understanding of how cyberinfrastructures emerge over time through processes of stakeholder alignment, enrollment, and through synergies among stakeholder groups. We explore these issues to better understand how cyberinfrastructures can be designed to sustain over time.
     2341</li>
     2342<br>
     2343
     2344
     2345
     2346<li>
     2347<b>Ravi, Abhiram and Ramanathan, Parmesh and Sivalingam, KrishnaM</b>
     2348, &quot;Integrated network coding and caching in information-centric networks: revisiting pervasive caching in the ICN framework.&quot;
     2349Photonic Network Communications, Springer US,
     23502015.
     2351doi:10.1007/s11107-015-0557-4.
     2352<a href="http://dx.doi.org/10.1007/s11107-015-0557-4">http://dx.doi.org/10.1007/s11107-015-0557-4</a>
     2353<br><br><b>Abstract: </b>Information-centric networks (ICNs) replace IP addresses with content names at the thin waist of the Internet hourglass, thereby enabling pervasive router-level caching at the network layer. In this paper, we revisit pervasive content caching and propose an algorithm for cache replacement at ICN routers by incorporating principles from network cod- ing, a technique used to achieve maximum flow rates in multicast. By introducing a low computational cost in the system, network-coded caching better utilizes the available small storage space at the routers to cache more effectively in the network. Results of our experiments on the global enterprise for network innovations (GENI) testbed demon- strating the performance of our algorithm on a real network are included in the paper. We evaluate the algorithm in two different traffic scenarios (i) video-on-demand (VoD) (ii) Zipf-based web traffic. Working with the named data networking implementation of ICN, we also present the addi- tional headers and logical components that are needed to enable network-coded caching. In a nutshell, we show that an integrated coding-and-caching strategy can provide sig- nificant gains in latency and content delivery rate for a small computational overhead.
     2354</li>
     2355<br>
     2356
     2357
     2358
     2359<li>
     2360<b>Raychaudhuri, Dipankar and Nagaraja, Kiran and Venkataramani, Arun</b>
     2361, &quot;MobilityFirst: a robust and trustworthy mobility-centric architecture for the future internet.&quot;
     2362SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
     23632012.
     2364doi:10.1145/2412096.2412098.
     2365<a href="http://dx.doi.org/10.1145/2412096.2412098">http://dx.doi.org/10.1145/2412096.2412098</a>
     2366<br><br><b>Abstract: </b>This paper presents an overview of the MobilityFirst network architecture, currently under development as part of the US National Science Foundation's Future Internet Architecture (FIA) program. The proposed architecture is intended to directly address the challenges of wireless access and mobility at scale, while also providing new services needed for emerging mobile Internet application scenarios. After briefly outlining the original design goals of the project, we provide a discussion of the main architectural concepts behind the network design, identifying key features such as separation of names from addresses, public-key based globally unique identifiers (GUIDs) for named objects, global name resolution service (GNRS) for dynamic binding of names to addresses, storage-aware routing and late binding, content- and context-aware services, optional in-network compute layer, and so on. This is followed by a brief description of the MobilityFirst protocol stack as a whole, along with an explanation of how the protocol works at end-user devices and inside network routers. Example of specific advanced services supported by the protocol stack, including multi-homing, mobility with disconnection, and content retrieval/caching are given for illustration. Further design details of two key protocol components, the GNRS name resolution service and the GSTAR routing protocol, are also described along with sample results from evaluation. In conclusion, a brief description of an ongoing multi-site experimental proof-of-concept deployment of the MobilityFirst protocol stack on the GENI testbed is provided.
     2367</li>
     2368<br>
     2369
     2370
     2371
     2372<li>
     2373<b>Ricart, Glenn</b>
     2374, &quot;US Ignite testbeds: Advanced testbeds enable next-generation applications.&quot;
     2375Teletraffic Congress (ITC), 2014 26th International, IEEE,
     23762014.
     2377doi:10.1109/itc.2014.6932975.
     2378<a href="http://dx.doi.org/10.1109/itc.2014.6932975">http://dx.doi.org/10.1109/itc.2014.6932975</a>
     2379<br><br><b>Abstract: </b>US Ignite is organizing what will eventually become 200 testbeds for next-generation applications in the United States. Twenty-eight testbeds are currently in various stages of operation. Most testbeds have gigabit to the end user capability including homes and small businesses. Both wired (fiber) and wireless cities are represented. The three salient advantages of these testbeds are their (1) applicability for big data (and big video) applications upstream and downstream, (2) ability to provide low-latency access to edge or local cloud (locavore) infrastructure for ultra-responsive and powerful applications, and (3) capacity for enough physical bandwidth to allow for virtualized channels carrying new services under new business models.
     2380</li>
     2381<br>
     2382
     2383
     2384
     2385<li>
     2386<b>Ricci, Robert and Eide, Eric</b>
     2387, &quot;Introducing CloudLab:Scientific Infrastructure for Advancing Cloud Architecturesand Applications.&quot;
     2388;login:, Usenix,
     23892014.
     2390
     2391<a href="http://www.usenix.org/publications/login/dec14/ricci">http://www.usenix.org/publications/login/dec14/ricci</a>
     2392<br><br><b>Abstract: </b>Researchers and practitioners are flush with ideas for tomorrow's cloud architectures. Their proposals range from small extensions of today's popular cloud-software stacks to all-new architectures that address mobility, energy efficiency, security and privacy, spe- cific workloads, the Internet of Things, and on and on. Many of the ideas that drive modern clouds, such as virtualization, network slicing, and robust distributed storage arose from the research community. However, today's clouds have become unsuitable for moving this research agenda forward: they have specific, unmalleable implementations of the core tech- nologies ” baked in.” To support next-generation cloud research, the community needs infrastructure that is built to support research into a wide variety of cloud architectures. CloudLab is a new, large-scale, diverse, and distributed infrastructure designed to address this need. CloudLab is not itself a cloud. Rather, it is a substrate on which researchers can build their own clouds and experi- ment with them in an environment that provides a high degree of realism.
     2393</li>
     2394<br>
     2395
     2396
     2397
     2398<li>
     2399<b>Ricci, Robert and Wong, Gary and Stoller, Leigh and Duerig, Jonathon</b>
     2400, &quot;An Architecture For International Federation of Network Testbeds.&quot;
     2401IEICE Transactions on Communications,
     24022013.
     2403doi:10.1587/transcom.E96.B.2.
     2404<a href="http://dx.doi.org/10.1587/transcom.E96.B.2">http://dx.doi.org/10.1587/transcom.E96.B.2</a>
     2405<br><br><b>Abstract: </b>Testbeds play a key role in the advancement of network science and the exploration of new network architectures. Because the scale and scope of any individual testbed is necessarily limited, federation is a useful technique for constructing testbeds that serve a wide range of experimenter needs. In a federated testbed, individual facilities maintain local autonomy while cooperating to provide a unified set of abstractions and interfaces to users. Forming an international federation is particularly challenging, because issues of trust, user access policy, and local laws and regulations are of greater concern that they are for federations within a single country. In this paper, we describe an architecture, based on the US National Science Foundation's GENI project, that is capable of supporting the needs of an international federation.
     2406</li>
     2407<br>
     2408
     2409
     2410
     2411<li>
     2412<b>Ricci, Robert and Wong, Gary and Stoller, Leigh and Webb, Kirk and Duerig, Jonathon and Downie, Keith and Hibler, Mike</b>
     2413, &quot;Apt: A Platform for Repeatable Research in Computer Science.&quot;
     2414SIGOPS Oper. Syst. Rev., ACM, New York, NY, USA,
     24152015.
     2416doi:10.1145/2723872.2723885.
     2417<a href="http://dx.doi.org/10.1145/2723872.2723885">http://dx.doi.org/10.1145/2723872.2723885</a>
     2418<br><br><b>Abstract: </b>Repeating research in computer science requires more than just code and data: it requires an appropriate environment in which to run experiments. In some cases, this environment appears fairly straightforward: it consists of a particular operating system and set of required libraries. In many cases, however, it is considerably more complex: the execution environment may be an entire network, may involve complex and fragile configuration of the dependencies, or may require large amounts of resources in terms of computation cycles, network bandwidth, or storage. Even the &#x73;&#x0308;traightforward&#x20;&#x0308;case turns out to be surprisingly intricate: there may be explicit or hidden dependencies on compilers, kernel quirks, details of the ISA, etc. The result is that when one tries to repeat published results, creating an environment sufficiently similar to one in which the experiment was originally run can be troublesome; this problem only gets worse as time passes. What the computer science community needs, then, are environments that have the explicit goal of enabling repeatable research. This paper outlines the problem of repeatable research environments, presents a set of requirements for such environments, and describes one facility that attempts to address them.
     2419</li>
     2420<br>
     2421
     2422
     2423
     2424<li>
     2425<b>Riga, Niky and Thomas, Vicraj and Maglaris, Vasilis and Grammatikou, Mary and Anifantis, Evangelos</b>
     2426, &quot;Virtual Laboratories - Use of Public Testbeds in Education.&quot;
     2427Proceedings of the 7th International Conference on Computer Supported Education, Lisbon, Portugal, SCITEPRESS - Science and and Technology Publications,
     24282015.
     2429doi:10.5220/0005496105160521.
     2430<a href="http://dx.doi.org/10.5220/0005496105160521">http://dx.doi.org/10.5220/0005496105160521</a>
     2431<br><br><b>Abstract: </b>Experimentation is an invaluable part of learning in all sciences. However, building and maintaining laboratories is expensive, time and space consuming. Moreover, in computer science advances in technology can quickly make the infrastructure obsolete. In this paper we advocate the use of recently deployed public testbeds as remote labs for computer science education. As an example we describe the successful use of the GENI testbed in graduate and undergraduate courses and present a specific case study of GENI being used in an undergraduate class on Network Management and Intelligent Networks.
     2432</li>
     2433<br>
     2434
     2435
     2436
     2437<li>
     2438<b>Risdianto, Aris C. and Kim, JongWon</b>
     2439, &quot;Prototyping Media Distribution Experiments over OF@TEIN SDN-enabled Testbed.&quot;
     2440Proceedings of the Asia-Pacific Advanced Network,
     24412014.
     2442doi:10.7125/apan.38.2.
     2443<a href="http://dx.doi.org/10.7125/apan.38.2">http://dx.doi.org/10.7125/apan.38.2</a>
     2444<br><br><b>Abstract: </b>The lifecycle of service realization experiment is composed of multiple stages, where tasks and responsibilities are well-defined between users and operators. In this paper, we prototype media distribution experiments over an OF@TEIN SDN (Software-Defined Networking)-enabled testbed while paying attention to the automated resource provisioning and experiment execution.
     2445</li>
     2446<br>
     2447
     2448
     2449
     2450<li>
     2451<b>Rivera and Fei, Zongming and Griffioen, James</b>
     2452, &quot;Providing a High Level Abstraction for SDN Networks in GENI.&quot;
     2453Distributed Computing Systems Workshops (ICDCSW), 2015 IEEE 35th International Conference on, IEEE,
     24542015.
     2455doi:10.1109/icdcsw.2015.22.
     2456<a href="http://dx.doi.org/10.1109/icdcsw.2015.22">http://dx.doi.org/10.1109/icdcsw.2015.22</a>
     2457<br><br><b>Abstract: </b>Software Defined Networks make it possible to decouple routing from forwarding, allowing the routing decisions to be made by a (logically)centralized controller which are then communicated to the switches in the network (for example, via the Open Flow protocol). One problem facing end users is the need to map high level abstractions -- like the path a flow should take -- to a set of low level forwarding rules tailored to, and installed at, every switch along the path. Installing such rules manually is tedious and error prone, and writing a controller program to do it is equally, if not more, challenging. In this paper, we propose a new set of tools that allow users (experimenters)to easily map their high level routing policies to low level Open Flow rules, and to help users reverse engineer high level policies from the installed set of low level flow rules. The tools provide users with the abstraction of end-to-end flows that users can install, list, and delete. The tools automatically handle the details of computing and installing all the rules needed to implement end-to-end flows, and are also capable of identifying flows and, if desired, removing flows that already exist. The tools have been implemented as modules in the GENI Desktop providing users with a graphical interface to their flows. In addition, we have implemented a module to monitor the performance of flows that have been installed. We describe our prototype implementation and present performance numbers obtained via the service.
     2458</li>
     2459<br>
     2460
     2461
     2462
     2463<li>
     2464<b>Rohrer, Justin P. and &#x43;&#x0327;etinkaya, Egemen K. and Sterbenz, James P. G.</b>
     2465, &quot;Progress and challenges in large-scale future internet experimentation using the GpENI programmable testbed.&quot;
     2466Proceedings of the 6th International Conference on Future Internet Technologies, Seoul, Republic of Korea, ACM, New York, NY, USA,
     24672011.
     2468doi:10.1145/2002396.2002409.
     2469<a href="http://dx.doi.org/10.1145/2002396.2002409">http://dx.doi.org/10.1145/2002396.2002409</a>
     2470<br><br><b>Abstract: </b>GpENI is evolving to provide a promising environment in which to do experimental research in the resilience and survivability of future networks, by allowing programmable control over topology and mechanism, while providing the scale and global reach needed to conduct network experiments far beyond the capabilities of a conventional testbed. Addressing this need at scale introduces a number of challenges both in deployment and in collecting results that can be directly compared to simulation results for cross-verification purposes. In this short paper we present the scope, design goals, challenges, and current status of the GpENI programmable testbed, as well as an overview and examples of the types of experiments we are beginning to run.
     2471</li>
     2472<br>
     2473
     2474
     2475
     2476<li>
     2477<b>Rosen, Aaron</b>
     2478, &quot;Network Service Delivery and Throughput Optimization via Software Defined Networking (Master's Thesis).&quot;
     2479
     24802012.
     2481
     2482<a href="http://tigerprints.clemson.edu/all&#x005F;theses/1332/">http://tigerprints.clemson.edu/all&#x005F;theses/1332/</a>
     2483<br><br><b>Abstract: </b>In today's world, transmitting data across large bandwidth-delay product (BDP) networks requires special configuration on end users' machines in order to be done efficiently. This added level of complexity creates extra cost and is usually overlooked by users unknowledgeable to the issues. This is one example problem which can be ameliorated with the emerging software defined networking (SDN) paradigm. In an SDN, packet forwarding is controlled via software controllers. In an OpenFlow SDN, a controller can control the forwarding, rewriting, and dropping of packets based on their header attributes. The ability to handle packets in customizable ways in software has significant implications for both users and operators of the network. Via SDN, network providers can easily provide services to enhance users' experience of the network. Steroid OpenFlow Service (SOS) is presented as a solution to seamless enhancement of TCP data transfer throughput over large BDP networks without any modification to the software and configurations on users' machines. SOS utilizes OpenFlow to redirect application specific traffic to application specific service agents. SOS uses service agents on both ends of the connection to seamlessly terminate a user's TCP connection, launch a set of parallel TCP connections, and leverage multiple paths when available to maximize throughput.
     2484</li>
     2485<br>
     2486
     2487
     2488
     2489<li>
     2490<b>Rosen, Aaron and Wang, Kuang-Ching</b>
     2491, &quot;Steroid OpenFlow Service: Seamless Network Service Delivery in Software Defined Networks.&quot;
     2492First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     24932012.
     2494
     2495
     2496<br><br><b>Abstract: </b>In a software defined network (SDN), packet forwarding is controlled by software controllers. In an OpenFlow SDN, a controller can control the forwarding, rewriting, and dropping of packets based on their header attributes. The ability to handle packets in customizable ways in software has significant implications for both network users and operators. Via software, users can convey application specific expectations while operators can deliver application specific services to enhance user experiences. In this paper, we present the Steroid OpenFlow Services (SOS) paradigm for network services delivery. The paradigm enables operators to deliver network services without any setup requirements on user machines. SOS utilizes OpenFlow to redirect application specific traffic to application specific service agents; SOS also rewrites packet headers for a service to remain seamless to users. This paper presents an example SOS service for optimizing large volume TCP download across a large delay-bandwidth-product wide area network. SOS service agents on both ends of the connection seamlessly terminate a user TCP connection, launch a set of parallel TCP connections, and leverage multiple paths when available to maximize throughput. With the NSF GENI future Internet testbed, a prototype implementation achieved up to 320 times throughput enhancement seamless to the end users.
     2497</li>
     2498<br>
     2499
     2500
     2501
     2502<li>
     2503<b>Ruth, Paul and Mandal, Anirban</b>
     2504, &quot;Domain Science Applications on GENI: Presentation and Demo.&quot;
     2505Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     25062014.
     2507doi:10.1109/icnp.2014.86.
     2508<a href="http://dx.doi.org/10.1109/icnp.2014.86">http://dx.doi.org/10.1109/icnp.2014.86</a>
     2509<br><br><b>Abstract: </b>Multi-tenant cloud infrastructures are increasingly used for high-performance and high-throughput domain science applications. In recent years, machine virtualization has come a long way toward supporting domain science applications. Various cloud platforms, such as Open Stack, Cloud Stack, and Amazon EC2 are attracting scientists to these platforms with the promise of customized environments with virtually infinite compute resources. At the same time, research efforts, such as NSF GENI are bringing together cloud computing with advanced network infrastructure provisioning. This paper presents work toward evaluating the use of GENI to support domain science applications. The evaluation involved two different domain science applications deployed on ExoGENI and Insta GENI. The first application is ADCIRC, a storm surge model that uses Message Passing Interface (MPI). The second is Motif network, a genomics application using the Pegasus workflow management system to manage a large data-intensive workflow.
     2510</li>
     2511<br>
     2512
     2513
     2514
     2515<li>
     2516<b>Ruth, Paul and Mandal, Anirban and Castillo, Claris and Fowler, Robert and Tilson, Jeff and Baldin, Ilya and Xin, Yufeng</b>
     2517, &quot;Achieving Performance Isolation on Multi-Tenant Networked Clouds Using Advanced Block Storage Mechanisms.&quot;
     2518Proceedings of the 6th Workshop on Scientific Cloud Computing, Portland, Oregon, USA, ACM, New York, NY, USA,
     25192015.
     2520doi:10.1145/2755644.2755649.
     2521<a href="http://dx.doi.org/10.1145/2755644.2755649">http://dx.doi.org/10.1145/2755644.2755649</a>
     2522<br><br><b>Abstract: </b>Multi-tenant cloud infrastructures are increasingly used for high-performance and high-throughput domain science applications. Various cloud platforms, such as OpenStack and Amazon EC2, along with research efforts, such as NSF GENI and FutureGrid have attracted scientists to these platforms with the promise of virtually infinite compute resources. This paper presents work toward providing better resource allocation accounting in multi-tenant cloud environments by understanding the subtle interference between network, compute, and storage resources. The experiments provide insight that help cloud administrators know how to best distribute virtual cores to physical cores considering the effect of advanced virtual network technologies on remote block I/O performance. The results show that SR-IOV network interfaces to an SSD iSCSI device can provide extremely fast block I/O with minimal CPU overhead and minimal performance interference between tenants. In addition, careful mapping of virtual computation to physical computational cores is critical to increasing performance isolation.
     2523</li>
     2524<br>
     2525
     2526
     2527
     2528<li>
     2529<b>Schlinker, Brandon and Zarifis, Kyriakos and Cunha, Italo and Feamster, Nick and Katz-Bassett, Ethan</b>
     2530, &quot;PEERING: An AS for Us.&quot;
     2531Proceedings of the 13th ACM Workshop on Hot Topics in Networks, Los Angeles, CA, USA, ACM, New York, NY, USA,
     25322014.
     2533doi:10.1145/2670518.2673887.
     2534<a href="http://dx.doi.org/10.1145/2670518.2673887">http://dx.doi.org/10.1145/2670518.2673887</a>
     2535<br><br><b>Abstract: </b>Internet routing suffers from persistent and transient failures, circuitous routes, oscillations, and prefix hijacks. A major impediment to progress is the lack of ways to conduct impactful interdomain research. Most research is based either on passive observation of existing routes, keeping researchers from assessing how the Internet will respond to route or policy changes; or simulations, which are restricted by limitations in our understanding of topology and policy. We propose a new class of interdomain research: researchers can instantiate an AS of their choice, including its intradomain topology and interdomain interconnectivity, and connect it with the &#x6c;&#x0308;ive&#x20;&#x0308;Internet to exchange routes and traffic with real interdomain neighbors. Instead of being observers of the Internet ecosystem, researchers become members. Towards this end, we present the Peering testbed. In its nascent stage, the testbed has proven extremely useful, resulting in a series of studies that were nearly impossible for researchers to conduct in the past. In this paper, we present a vision of what the testbed can provide. We sketch how to extend the testbed to enable future innovation, taking advantage of the rise of IXPs to expand our testbed.
     2536</li>
     2537<br>
     2538
     2539
     2540
     2541<li>
     2542<b>Schwerdel, Dennis and Reuther, Bernd and Zinner, Thomas and M&#x75;&#x0308;ller, Paul and Tran-Gia, Phouc</b>
     2543, &quot;Future Internet research and experimentation: The G-Lab approach.&quot;
     2544Computer Networks,
     25452014.
     2546doi:10.1016/j.bjp.2013.12.023.
     2547<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.023">http://dx.doi.org/10.1016/j.bjp.2013.12.023</a>
     2548<br><br><b>Abstract: </b>The German Lab (G-Lab) project aims to investigate architectural concepts and technologies for a new inter-networking architecture as an integrated approach between theoretic and experimental studies. Thus G-Lab consists of two major fields of activities: research studies of future network components and the design and setup of experimental facilities. Both are controlled by the same community to ensure that the experimental facility meets the demands of the researchers. Researchers gain access to virtualized resources or may gain exclusive access to resources if necessary. We present the current setup of the experimental facility, describing the available hardware, management of the platform, the utilization of the PlanetLab software and the user management. Moreover, a new approach to setup and deploy virtual network topologies will be described.
     2549</li>
     2550<br>
     2551
     2552
     2553
     2554<li>
     2555<b>Scoglio, Caterina M. and Sydney, Ali and Youssef, Mina and Schumm, Phillip and Kooij, Robert E.</b>
     2556, &quot;Elasticity and Viral Conductance: Unveiling Robustness in Complex Networks through Topological Characteristics.&quot;
     2557CoRR,
     25582008.
     2559
     2560<a href="http://arxiv-web3.library.cornell.edu/abs/0811.3272v3">http://arxiv-web3.library.cornell.edu/abs/0811.3272v3</a>
     2561<br><br><b>Abstract: </b>With increasingly ambitious initiatives such as GENI and FIND that seek to design the future Internet, it becomes imperative to define the characteristics of robust topologies, and build future networks optimized for robustness. This paper investigates the characteristics of network topologies that maintain a high level of throughput in spite of multiple attacks. To this end, we select network topologies belonging to the main network models and some real world networks. We consider three types of attacks: removal of random nodes, high degree nodes, and high betweenness nodes. We use elasticity as our robustness measure and, through our analysis, illustrate that different topologies can have different degrees of robustness. In particular, elasticity can fall as low as 0.8&#x0025; of the upper bound based on the attack employed. This result substantiates the need for optimized network topology design. Furthermore, we implement a tradeoff function that combines elasticity under the three attack strategies and considers the cost of the network. Our extensive simulations show that, for a given network density, regular and semi-regular topologies can have higher degrees of robustness than heterogeneous topologies, and that link redundancy is a sufficient but not necessary condition for robustness.
     2562</li>
     2563<br>
     2564
     2565
     2566
     2567<li>
     2568<b>Seetharam, Sripriya</b>
     2569, &quot;Application-Driven Overlay Network as a Service for Data-Intensive Science (Master's thesis).&quot;
     2570
     25712014.
     2572
     2573<a href="https://mospace.umsystem.edu/xmlui/handle/10355/45798">https://mospace.umsystem.edu/xmlui/handle/10355/45798</a>
     2574<br><br><b>Abstract: </b>Campuses are increasingly adopting hybrid cloud architectures for supporting data-intensive science applications that require &#x6f;&#x0308;n-demand&#x20;&#x0308;resources, which are not always available locally on-site. Policies at the campus edge for handling multiple such applications competing for remote resources can cause bottlenecks across applications. These bottlenecks can be proactively avoided with pertinent profiling, monitoring and control of application flows using the emerging paradigm of software-defined networking (SDN). In this thesis, we leverage SDN principles in the design and implementation of an &#x41;&#x0308;pplication-driven Overlay Network-as-a-Service&#x20;&#x0308;(ADON) framework that can manage the hybrid cloud requirements of multiple applications in a scalable and extensible manner. ADON's main features include: programmable &#x63;&#x0308;ustom templates&#x20;&#x0308;and a &#x76;&#x0308;irtual tenant handler&#x20;&#x0308;algorithm. Our solution approach in ADON involves scheduling transit selection and traffic engineering at the campus-edge based on real-time policy control that ensures predictable application performance delivery for multi-tenant traffic profiles. We also present a market-driven (distributed) resource optimization scheme that can address the Internet-scale scalability problems of handling resource requests of multiple data-intensive applications within a desktop-as-a-service cloud environment. We show how our optimization scheme can increase the system performance and user experience levels using metrics such as 'Service Response Time' and 'Net-Utility'. Lastly,we discuss ADON effectiveness validation with an implementation on a wide-area overlay network testbed featuring temporal behavior of multi-tenant traffic burst arrivals. We conclude by presenting hybrid cloud implementation best practices that ease the orchestration of network programmability for campus network providers and data-intensive application users.
     2575</li>
     2576<br>
     2577
     2578
     2579
     2580<li>
     2581<b>Seetharam, Sripriya and Calyam, Prasad and Beyene, Tsegereda</b>
     2582, &quot;ADON: Application-Driven Overlay Network-as-a-Service for Data-Intensive Science.&quot;
     2583
     25842014.
     2585doi:10.1109/CloudNet.2014.6969014.
     2586<a href="http://people.cs.missouri.edu/&#x63;&#x0303;alyamp/publications/adon-cloudnet14.pdf">http://people.cs.missouri.edu/&#x63;&#x0303;alyamp/publications/adon-cloudnet14.pdf</a>
     2587<br><br><b>Abstract: </b>Campuses are increasingly adopting hybrid cloud architectures for supporting data-intensive science applications that require ” on-demand” resources, which are not always available locally on-site. Policies at the campus edge for handling multiple such applications competing for remote resources can cause bottlenecks across applications. These bottlenecks can be proactively avoided with pertinent profiling, monitoring and control of application flows using software-defined networking principles. In this paper, we present an ” Application-driven Overlay Network-as-a-Service” (ADON) that can manage the hybrid cloud requirements of multiple applications in a scalable and extensible manner using features such as: programmable ” custom templates” and a ” virtual tenant handler”. Our solution approach involves scheduling transit selection and traffic engi- neering at the campus-edge based on real-time policy control that ensures predictable application performance delivery for multi-tenant traffic profiles. We validate our ADON approach with an implementation on a wide-area overlay network testbed across two campuses, and present a workflow that eases the orchestration of network programmability for campus network providers and data-intensive application users. Lastly, we present an emulation study of the ADON effectiveness in handling temporal behavior of multi-tenant traffic burst arrivals using profiles from a diverse set of actual data-intensive applications.
     2588</li>
     2589<br>
     2590
     2591
     2592
     2593<li>
     2594<b>Selvadhurai, Arunprasaath</b>
     2595, &quot;Network Measurement Tool Components for Enabling Performance Intelligence within Cloud-based Applications (Master's Thesis).&quot;
     2596
     25972013.
     2598
     2599<a href="http://rave.ohiolink.edu/etdc/view?acc&#x005F;num=osu1367446588">http://rave.ohiolink.edu/etdc/view?acc&#x005F;num=osu1367446588</a>
     2600<br><br><b>Abstract: </b>Popular applications such as email, photo/video galleries, and file storage are increasingly being supported by cloud platforms in residential, academic and industry communities. The next frontier for these user communities will be to transition `traditional desktops' that have dedicated hardware and software configurations into `virtual desktop clouds' that are accessible via thin-clients. In our thesis, we show how the underlying measurement services, with some additional capabilities, can be used as intelligent agents to provide network intelligence within thin-client based virtual desktops applications. The framework leverages principles of software defined networking and features an `unified resource broker' that uses special `marker packets' for: (a) ” route setup” when handling non-IP traffic between thin-client sites and data centers, (b) ” path selection” and ” load balancing” of virtual desktop flows to improve the performance of interactive applications and video playback, and to cope with faults such as link-failures or Denial-of-Service cyber-attacks. In addition, we detail our framework implementation within a virtual desktop cloud (VDC) in a multi-domain Global Environment for Network Innovations (GENI). We present empirical results from our experimentation that leverages OpenFlow programmable networking, as well as OnTimeMeasure instrumentation-and-measurement capabilities for validating our framework in GENI under realistic settings. Our results demonstrate the importance of scheduling regulated measurements that can be used for intelligent resource placement decisions. Our results also show the feasibility and benefits of using the measurement services for effective path selection and load balancing between thin-client sites and data centers in VDCs and simulation applications.
     2601</li>
     2602<br>
     2603
     2604
     2605
     2606<li>
     2607<b>Seskar, Ivan and Nagaraja, Kiran and Nelson, Sam and Raychaudhuri, Dipankar</b>
     2608, &quot;MobilityFirst future internet architecture project.&quot;
     2609Proceedings of the 7th Asian Internet Engineering Conference, Bangkok, Thailand, ACM, New York, NY, USA,
     26102011.
     2611doi:10.1145/2089016.2089017.
     2612<a href="http://dx.doi.org/10.1145/2089016.2089017">http://dx.doi.org/10.1145/2089016.2089017</a>
     2613<br><br><b>Abstract: </b>This short paper presents an overview of the MobilityFirst network architecture, which is a clean-slate project being conducted as part of the NSF Future Internet Architecture (FIA) program. The proposed architecture is intended to directly address the challenges of wireless access and mobility at scale, while also providing new multicast, anycast, multi-path and context-aware services needed for emerging mobile Internet application scenarios. Key protocol components of the proposed architecture are: (a) separation of naming from addressing; (b) public key based self-certifying names (called globally unique identifiers or GUIDs) for network-attached objects; (c) global name resolution service (GNRS) for dynamic name-to-address binding; (d) delay-tolerant and storage-aware routing (GSTAR) capable of dealing with wireless link quality fluctuations and disconnections; (e) hop-by-hop transport of large protocol data units; and (f) location or context-aware services. The basic operations of a MobilityFirst router are outlined. This is followed by a discussion of ongoing proof-of-concept prototyping and experimental evaluation efforts for the MobilityFirst protocol stack. In conclusion, a brief description of an ongoing multi-site experimental deployment of the MobilityFirst protocol stack on the GENI testbed is provided.
     2614</li>
     2615<br>
     2616
     2617
     2618
     2619<li>
     2620<b>Sharma, Navin and Gummeson, Jeremy and Irwin, David and Shenoy, Prashant</b>
     2621, &quot;Cloudy Computing: Leveraging Weather Forecasts in Energy Harvesting Sensor Systems.&quot;
     26222010 7th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON), Boston, MA, USA, IEEE,
     26232010.
     2624doi:10.1109/SECON.2010.5508260.
     2625<a href="http://dx.doi.org/10.1109/SECON.2010.5508260">http://dx.doi.org/10.1109/SECON.2010.5508260</a>
     2626<br><br><b>Abstract: </b>To sustain perpetual operation, systems that harvest environmental energy must carefully regulate their usage to satisfy their demand. Regulating energy usage is challenging if a system's demands are not elastic and its hardware components are not energy-proportional, since it cannot precisely scale its usage to match its supply. Instead, the system must choose when to satisfy its energy demands based on its current energy reserves and predictions of its future energy supply. In this paper, we explore the use of weather forecasts to improve a system's ability to satisfy demand by improving its predictions. We analyze weather forecast, observational, and energy harvesting data to formulate a model that translates a weather forecast to a wind or solar energy harvesting prediction, and quantify its accuracy. We evaluate our model for both energy sources in the context of two different energy harvesting sensor systems with inelastic demands: a sensor testbed that leases sensors to external users and a lexicographically fair sensor network that maintains steady node sensing rates. We show that using weather forecasts in both wind- and solar-powered sensor systems increases each system's ability to satisfy its demands compared with existing prediction strategies.
     2627</li>
     2628<br>
     2629
     2630
     2631
     2632<li>
     2633<b>Shen, Haiying and Liu, Guoxin</b>
     2634, &quot;Harmony: Integrated Resource and Reputation Management for Large-Scale Distributed Systems.&quot;
     26352011 Proceedings of 20th International Conference on Computer Communications and Networks (ICCCN), Lahaina, HI, USA, IEEE,
     26362011.
     2637doi:10.1109/ICCCN.2011.6005739.
     2638<a href="http://dx.doi.org/10.1109/ICCCN.2011.6005739">http://dx.doi.org/10.1109/ICCCN.2011.6005739</a>
     2639<br><br><b>Abstract: </b>Advancements in technology over the past decade are leading to a promising future for large-scale distributed systems, where globally-scattered distributed resources are collectively pooled and used in a cooperative manner to achieve unprecedented petascale supercomputing capabilities. The issues of resource management (resMgt) and reputation management (repMgt) need to be addressed in order to ensure the successful deployment of large-scale distributed systems. However, these two issues have typically been addressed separately, despite the significant interdependencies between them: resMgt needs repMgt to provide a cooperative environment for resource sharing, and in turn facilitates repMgt to evaluate multi-faceted node reputations for providing different resources. Current repMgt methods provide a single reputation value for each node in providing all types of resources. However, a node willing to provide one resource may not be willing to provide another resource. In addition, current repMgt methods often guide node selection policy to select the highest-reputed nodes, which may overload these nodes. Also, few works exploited node reputation in resource selection in order to fully and fairly utilize resources in the system and to meet users' diverse QoS demands. We propose a system called Harmony that integrates resMgt and repMgt in a harmonious manner. Harmony incorporates two key innovations: integrated multi-faceted resource/reputation management and multi-QoS-oriented resource selection. The trace data we collected from an online trading platform confirms the importance of multi-faceted reputation and potential problems with highest-reputed node selection. Trace-driven experiments performed on PlanetLab show that Harmony outperforms existing resMgt and repMgt in terms of the success rate, service delay, and efficiency.
     2640</li>
     2641<br>
     2642
     2643
     2644
     2645<li>
     2646<b>Sher-DeCusatis, Carolyn J. and DeCusatis, Casimer</b>
     2647, &quot;Developing a Software Defined Networking curriculum through industry partnerships.&quot;
     2648American Society for Engineering Education (ASEE Zone 1), 2014 Zone 1 Conference of the,
     26492014.
     2650doi:10.1109/ASEEZone1.2014.6820653.
     2651<a href="http://dx.doi.org/10.1109/ASEEZone1.2014.6820653">http://dx.doi.org/10.1109/ASEEZone1.2014.6820653</a>
     2652<br><br><b>Abstract: </b>Software Defined Networking (SDN) is an emerging technology which radically improves cloud computing and other types of data networking. We discuss a new SDN undergraduate education program, developed in collaboration with industry partnerships. Student labs using resources such as GENI, NetFPGA, and the New York State Cloud Computing Center will be presented. We also outline SDN student projects including firewalls, load balancers, and redundant failover systems.
     2653</li>
     2654<br>
     2655
     2656
     2657
     2658<li>
     2659<b>Shin, Sunae and Dhondge, Kaustubh and Choi, Baek-Young</b>
     2660, &quot;Understanding the Performance of TCP and UDP-based Data Transfer Protocols using EMULAB.&quot;
     2661First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     26622012.
     2663
     2664
     2665<br><br><b>Abstract: </b>In this paper, we present a hands-on course project that explores the performance of data transfer protocols using a GENI resource. TCP is one of the key topics in networking courses, and understanding its behavior as well as limitations, from real experiments, offers an invaluable and deep learning experience. A protocol's performance is directly impacted by network parameters such as network bandwidth, delay and loss. However, it is difficult to control and even vary those parameters, if it is not evaluated with simulations. GENI facilities conveniently provide a virtual laboratory that enables us to control the network settings with real network systems. Through this educational project, students had an opportunity to control important network parameters, and measure and compare TCP's performance with a UDP-based data transfer protocol, UDT, using EMULAB. Students were enthusiastic to witness the protocols' performances, and the limitations of TCP under a high bandwidth delay product network in the presence of packet loss, and to recognize the importance of protocol design and system issues for the future Internet.
     2666</li>
     2667<br>
     2668
     2669
     2670
     2671<li>
     2672<b>Singhal, Manav and Ramanathan, Jay and Calyam, Prasad and Skubic, Marjorie</b>
     2673, &quot;In-the-Know: Recommendation Framework for City-Supported Hybrid Cloud Services.&quot;
     2674Utility and Cloud Computing (UCC), 2014 IEEE/ACM 7th International Conference on, IEEE,
     26752014.
     2676doi:10.1109/ucc.2014.22.
     2677<a href="http://dx.doi.org/10.1109/ucc.2014.22">http://dx.doi.org/10.1109/ucc.2014.22</a>
     2678<br><br><b>Abstract: </b>Hybrid cloud architectures are particularly attractive to leverage city-level investments for building customized clouds, and for extending them to leverage public clouds. A successful design of the hybrid cloud architecture should facilitate the provisioning of scalable and secure services suited to a variety of communities such as residential homes and high-tech business incubators. In this paper, we present a novel &#x49;&#x0308;n-the-know&#x20;&#x0308;recommendation framework for provisioning of cloud resources in the form of 'on-demand contracts' to address the challenges in delivering the hybrid service variations for different community and individual needs. Our recommendation framework uses knowledge of the city's socio-economic goals/values as well as user preferences in terms of cost, performance and mobility. Using such knowledge, it recommends dynamic decisions by choosing from various provisioning alternatives in order to: (a) ensure optimal user Quality of Experience (QoE) in service delivery, and (b) effective utilization of hybrid cloud resources. We validate our recommendation framework using service composition experiments to satisfy an exemplar collaboration use case in an actual city-supported hybrid cloud test bed involving citizen consumers.
     2679</li>
     2680<br>
     2681
     2682
     2683
     2684<li>
     2685<b>Sivakumar, Ashiwan and Shankaranarayanan, P. N. and Rao, Sanjay</b>
     2686, &quot;Closer to the Cloud - A Case for Emulating Cloud Dynamics by Controlling the Environment.&quot;
     2687First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     26882012.
     2689
     2690
     2691
     2692</li>
     2693<br>
     2694
     2695
     2696
     2697<li>
     2698<b>Soroush, Hamed and Banerjee, Nilanjan and Corner, Mark and Levine, Brian and Lynn, Brian</b>
     2699, &quot;A retrospective look at the UMass DOME mobile testbed.&quot;
     2700SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
     27012012.
     2702doi:10.1145/2169077.2169079.
     2703<a href="http://dx.doi.org/10.1145/2169077.2169079">http://dx.doi.org/10.1145/2169077.2169079</a>
     2704<br><br><b>Abstract: </b>In this paper we describe the evolution of DOME, a diverse outdoor testbed for mobile experimentation. In addition, while highlighting the challenges faced in construction of DOME, we describe a concrete set of scientific results derived from this experience in a retrospective study. First, we argue that a broad range of mobility experiments could be performed in a testbed which provides the properties of temporal, technological, and spatial diversity. We demonstrate these properties in our testbed through analysis of data collected from DOME over a period of four years. Second, we crystallize a set of design principles that others should use when constructing testbeds of their own, including those related to deploying and managing a diverse testbed, distributing experiments remotely, and fostering collaborations among testbed stakeholders. Finally, using traces collected by DOME, we provide insights into several important problems in mobile systems research.
     2705</li>
     2706<br>
     2707
     2708
     2709
     2710<li>
     2711<b>Sridharan, Mukundan and Calyam, Prasad and Venkataraman, Aishwarya and Berryman, Alex</b>
     2712, &quot;Defragmentation of Resources in Virtual Desktop Clouds for Cost-Aware Utility-Optimal Allocation.&quot;
     27132011 Fourth IEEE International Conference on Utility and Cloud Computing, Melbourne, Australia, IEEE,
     27142011.
     2715doi:10.1109/UCC.2011.41.
     2716<a href="http://dx.doi.org/10.1109/UCC.2011.41">http://dx.doi.org/10.1109/UCC.2011.41</a>
     2717<br><br><b>Abstract: </b>Cloud Service Providers (CSPs) make virtual desktop cloud (VDC) resource provisioning decisions within desktop pools based on user groups and their application profiles. Such provisioning is aimed to satisfy acceptable user quality of experience (QoE) levels and is coupled with subsequent placement of VDs across distributed data centers. The placement decisions are influenced by session latency, load balancing and operation cost constraints. In this paper, we identify the resource fragmentation problem that occurs when placement is done opportunistically to minimize provisioning time and deliver satisfactory user QoE. To solve this problem, which inherently is an NP-Hard problem, we propose a defragmentation scheme that has fast convergence time and has three levels of complexity: (i) &#x75;&#x0308;tility fair provisioning&#x20;&#x0308;(UFP) to optimize resource provisioning within a data center - to achieve relative fairness between desktop pools, (ii) &#x73;&#x0308;tatic migration-free utility optimal placement and provisioning&#x20;&#x0308;(MUPP) to optimize resource provisioning between multiple data centers - to improve performance, and (iii) &#x64;&#x0308;ynamic global utility optimal placement and provisioning&#x20;&#x0308;(GUPP) to optimize resource provisioning using cost-aware and utility-maximal VD re-allocations and migrations - to increase scalability. We evaluate our defragmentation scheme against 'least latency', 'least load', and 'least cost' schemes using a novel &#x56;&#x0308;DC-Sim&#x20;&#x0308;simulator that we have developed in this study. Our simulations leverage profiles of user groups and their applications within desktop pools, obtained from a real VDC test bed. Our simulation results demonstrate that defragmentation is an important optimization step that can enable CSPs to achieve fairness, substantially improve user QoE and increase VDC scalability.
     2718</li>
     2719<br>
     2720
     2721
     2722
     2723<li>
     2724<b>Sridharan, Mukundan and Zeng, Wenjie and Leal, William and Ju, Xi and Ramanath, Rajiv and Zhang, Hongwei and Arora, Anish</b>
     2725, &quot;From Kansei to KanseiGenie: Architecture of Federated, Programmable Wireless Sensor Fabrics.&quot;
     2726Proceedings of the ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TridentCom),
     27272010.
     2728
     2729
     2730<br><br><b>Abstract: </b>This paper deals with challenges in federating wireless sensing fabrics. Federations of this sort are currently being developed in next generation global end-to-end experimentation infrastructures, such as GENI, to support rapid prototyping and hi-fidelity validation of protocols and applications. On one hand, federation should support access to diverse (and potentially provider-specific) wireless sensor resources and, on the other, it should enable users to uniformly task these resources. Instead of more simple basing federation upon a standard description of resources, we propose an architecture where the ontology of resource description can vary across providers, and a mapping of user needs to resources is performed to achieve uniform tasking. We illustrate one realization of this architecture, in terms of our refactoring the Kansei testbed to become the KanseiGenie federated fabric manager, which has full support for programmability, sliceability, and federated experimentation over heterogeneous sensing fabrics.
     2731</li>
     2732<br>
     2733
     2734
     2735
     2736<li>
     2737<b>Stabler, Greg and Goasguen, Sebastien and Rosen, Aaron and Wang, Kuang-Ching</b>
     2738, &quot;OneCloud: Controlling the Network in an OpenFlow Cloud.&quot;
     2739First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
     27402012.
     2741
     2742
     2743<br><br><b>Abstract: </b>Cloud computing is an emerging paradigm for on-demand access to computing resources over the network. Beyond early Software as a Service (SaaS) offerings, there is an increasing interest in the Infrastructure as a Service (IaaS) model where users request specific storage, networking, and computing resources to meet their application needs. To provision the network in a cloud, IaaS providers, such as the Amazon Web Services, allow users to choose their IP addresses, which can be associated with a dynamic set of virtual hosts (Elastic IP) with VPN, dynamic DNS, and dynamic firewall services. In this paper, we analyze a range of cloud network provisioning needs and the means to realize them in an OpenFlow network. We present an OpenFlow enabled framework for cloud network provisioning, based on the Open- Nebula cloud provisioning engine. Specifically, we demonstrate an Elastic IP service compatible with the Amazon Elastic Compute Cloud (EC2) API. This demonstration is available on the Clemson OneCloud IaaS offering. Ongoing efforts focus on the enablement of additional cloud network services for campus networks and wide area experimental networks like the National Science Foundation's GENI network.
     2744</li>
     2745<br>
     2746
     2747
     2748
     2749<li>
     2750<b>Stabler, Greg and Rosen, Aaron and Goasguen, Sebastien and Wang, Kuang-Ching</b>
     2751, &quot;Elastic IP and security groups implementation using OpenFlow.&quot;
     2752Proceedings of the 6th international workshop on Virtualization Technologies in Distributed Computing Date, Delft, The Netherlands, ACM, New York, NY, USA,
     27532012.
     2754doi:10.1145/2287056.2287069.
     2755<a href="http://doi.acm.org/10.1145/2287056.2287069">http://doi.acm.org/10.1145/2287056.2287069</a>
     2756<br><br><b>Abstract: </b>This paper presents a reference implementation of an Elastic IP and Security Group service using the OpenFlow protocol. The implementation is the first to present integration of OpenFlow within a virtual machine provisioning engine and an API for enabling such services. In this paper the OpenNebula system is used. The Elastic IP and Security Groups services are similar to the Amazon EC2 services and present a compatible Query API implemented by OpenNebula. The core of the implementation relies on the integration of an OpenFlow controller (NOX) with the EC2 server. Flow rules can be inserted in the OpenFlow controller using the EC2 API. These rules are then used by Open vSwitch bridges on the underlying hypervisor to manage network traffic. The reference implementation presented opens the door for more advanced cloud networking services that leverage principles from software defined networking including virtual private cloud, virtual data center spanning multiple availability zones, as well as seamless migration over wide are networks.
     2757</li>
     2758<br>
     2759
     2760
     2761
     2762<li>
     2763<b>Stavropoulos, Donatos and Dadoukis, Aris and Rakotoarivelo, Thierry and Ott, Max and Korakis, Thanasis and Tassiulas, Leandros</b>
     2764, &quot;Design, architecture and implementation of a resource discovery, reservation and provisioning framework for testbeds.&quot;
     2765Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt), 2015 13th International Symposium on, IEEE,
     27662015.
     2767doi:10.1109/wiopt.2015.7151032.
     2768<a href="http://dx.doi.org/10.1109/wiopt.2015.7151032">http://dx.doi.org/10.1109/wiopt.2015.7151032</a>
     2769<br><br><b>Abstract: </b>Experimental platforms (testbeds) play a significant role in the evaluation of new and existing technologies. Their popularity has been raised lately as more and more researchers prefer experimentation over simulation as a way for acquiring more accurate results. This imposes significant challenges in testbed operators since an efficient mechanism is needed to manage the testbed's resources and provision them according to the users' needs. In this paper we describe such a framework which was implemented for the management of networking testbeds. We present the design requirements and the implementation details, along with the challenges we encountered during its operation in the NITOS testbed. Significant results were extracted through the experiences of the every day operation of the testbed's management.
     2770</li>
     2771<br>
     2772
     2773
     2774
     2775<li>
     2776<b>Sterbenz, J. P. G. and Egemen and Hameed, M. A. and Jabbar, A. and Rohrer, J. P.</b>
     2777, &quot;Modelling and analysis of network resilience.&quot;
     27782011 Third International Conference on Communication Systems and Networks (COMSNETS 2011), Bangalore, IEEE,
     27792011.
     2780doi:10.1109/COMSNETS.2011.5716502.
     2781<a href="http://dx.doi.org/10.1109/COMSNETS.2011.5716502">http://dx.doi.org/10.1109/COMSNETS.2011.5716502</a>
     2782<br><br><b>Abstract: </b>As the Internet becomes increasingly important to all aspects of society, the consequences of disruption become increasingly severe. Thus it is critical to increase the resilience and survivability of the future network. We define resilience as the ability of the network to provide desired service even when challenged by attacks, large-scale disasters, and other failures. This paper describes a comprehensive methodology to evaluate network resilience using a combination of analytical and simulation techniques with the goal of improving the resilience and survivability of the Future Internet.
     2783</li>
     2784<br>
     2785
     2786
     2787
     2788<li>
     2789<b>Sterbenz, James P. G. and &#x43;&#x0327;etinkaya, Egemen K. and Hameed, Mahmood A. and Jabbar, Abdul and Qian, Shi and Rohrer, Justin P.</b>
     2790, &quot;Evaluation of network resilience, survivability, and disruption tolerance: analysis, topology generation, simulation, and experimentation.&quot;
     2791Telecommunication Systems, Telecommunication Systems, Springer US,
     27922013.
     2793doi:10.1007/s11235-011-9573-6.
     2794<a href="http://dx.doi.org/10.1007/s11235-011-9573-6">http://dx.doi.org/10.1007/s11235-011-9573-6</a>
     2795<br><br><b>Abstract: </b>As the Internet becomes increasingly important to all aspects of society, the consequences of disruption become increasingly severe. Thus it is critical to increase the resilience and survivability of future networks. We define resilience as the ability of the network to provide desired service even when challenged by attacks, large-scale disasters, and other failures. This paper describes a comprehensive methodology to evaluate network resilience using a combination of topology generation, analytical, simulation, and experimental emulation techniques with the goal of improving the resilience and survivability of the Future Internet.
     2796</li>
     2797<br>
     2798
     2799
     2800
     2801<li>
     2802<b>Su&#x6e;&#x0303;&#x65;&#x0301;, M. and Bergesio, L. and Woesner, H. and Rothe, T. and K&#x6f;&#x0308;psel, A. and Colle, D. and Puype, B. and Simeonidou, D. and Nejabati, R. and Channegowda, M. and Kind, M. and Dietz, T. and Autenrieth, A. and Kotronis, V. and Salvadori, E. and Salsano, S. and K&#x6f;&#x0308;rner, M. and Sharma, S.</b>
     2803, &quot;Design and implementation of the OFELIA FP7 facility: The European OpenFlow testbed.&quot;
     2804Computer Networks,
     28052014.
     2806doi:10.1016/j.bjp.2013.10.015.
     2807<a href="http://dx.doi.org/10.1016/j.bjp.2013.10.015">http://dx.doi.org/10.1016/j.bjp.2013.10.015</a>
     2808<br><br><b>Abstract: </b>The growth of the Internet in terms of number of devices, the number of networks associated to each device and the mobility of devices and users makes the operation and management of the Internet network infrastructure a very complex challenge. In order to address this challenge, innovative solutions and ideas must be tested and evaluated in real network environments and not only based on simulations or laboratory setups. OFELIA is an European FP7 project and its main objective is to address the aforementioned challenge by building and operating a multi-layer, multi-technology and geographically distributed Future Internet testbed facility, where the network itself is precisely controlled and programmed by the experimenter using the emerging OpenFlow technology. This paper reports on the work done during the first half of the project, the lessons learned as well as the key advantages of the OFELIA facility for developing and testing new networking ideas. An overview on the challenges that have been faced on the design and implementation of the testbed facility is described, including the OFELIA Control Framework testbed management software. In addition, early operational experience of the facility since it was opened to the general public, providing five different testbeds or islands, is described.
     2809</li>
     2810<br>
     2811
     2812
     2813
     2814<li>
     2815<b>Sun, Peng and Vanbever, Laurent and Rexford, Jennifer</b>
     2816, &quot;Scalable Programmable Inbound Traffic Engineering.&quot;
     2817Proceedings of the 1st ACM SIGCOMM Symposium on Software Defined Networking Research, Santa Clara, California, ACM, New York, NY, USA,
     28182015.
     2819doi:10.1145/2774993.2775063.
     2820<a href="http://dx.doi.org/10.1145/2774993.2775063">http://dx.doi.org/10.1145/2774993.2775063</a>
     2821<br><br><b>Abstract: </b>With the rise of video streaming and cloud services, enterprise and access networks receive much more traffic than they send, and must rely on the Internet to offer good end-to-end performance. These edge networks often connect to multiple ISPs for better performance and reliability, but have only limited ways to influence which of their ISPs carries the traffic for each service. In this paper, we present Sprite, a software-defined solution for flexible inbound traffic engineering (TE). Sprite offers direct, fine-grained control over inbound traffic, by announcing different public IP prefixes to each ISP, and performing source network address translation (SNAT) on outbound request traffic. Our design achieves scalability in both the data plane (by performing SNAT on edge switches close to the clients) and the control plane (by having local agents install the SNAT rules). The controller translates high-level TE objectives, based on client and server names, as well as performance metrics, to a dynamic network policy based on real-time traffic and performance measurements. We evaluate Sprite with live data from &#x69;&#x0308;n the wild&#x20;&#x0308;experiments on an EC2-based testbed, and demonstrate how Sprite dynamically adapts the network policy to achieve high-level TE objectives, such as balancing YouTube traffic among ISPs to improve video quality.
     2822</li>
     2823<br>
     2824
     2825
     2826
     2827<li>
     2828<b>Sydney, A. and Nutaro, J. and Scoglio, C. and Gruenbacher, D. and Schulz, N.</b>
     2829, &quot;Simulative Comparison of Multiprotocol Label Switching and OpenFlow Network Technologies for Transmission Operations.&quot;
     2830Smart Grid, IEEE Transactions on,
     28312013.
     2832doi:10.1109/TSG.2012.2227516.
     2833<a href="http://dx.doi.org/10.1109/TSG.2012.2227516">http://dx.doi.org/10.1109/TSG.2012.2227516</a>
     2834<br><br><b>Abstract: </b>Utility companies are integrating multiprotocol label switching (MPLS) technologies into existing backbone networks, including networks between substations and control centers. MPLS has mechanisms for efficient overlay technologies as well as mechanisms to enhance security: features essential to the functioning of the smart grid. However, with MPLS routing and other switching technologies innovation is restricted to the features enclosed ” in the box.” More specifically, there is no practical way for utility operators or researchers to test new ideas such as alternatives to IP or MPLS on a realistic scale to obtain the experience and confidence necessary for real world deployments. As a result, novel ideas go untested. Conversely, the OpenFlow framework has enabled significant advancements in network research. OpenFlow provides utility operators and researchers the programmability and flexibility necessary to enable innovation in next-generation communication architectures for the smart grid. This level of flexibility allows OpenFlow to provide all features of MPLS and also allows OpenFlow devices to co-exist with existing MPLS devices. The simulation results in this paper demonstrate that OpenFlow performs as well as MPLS, and may therefore be considered an alternative to MPLS for smart grid applications.
     2835</li>
     2836<br>
     2837
     2838
     2839
     2840<li>
     2841<b>Sydney, Ali</b>
     2842, &quot;The evaluation of software defined networking for communication and control of cyber physical systems (Doctoral dissertation).&quot;
     2843
     28442013.
     2845
     2846<a href="http://hdl.handle.net/2097/15577">http://hdl.handle.net/2097/15577</a>
     2847<br><br><b>Abstract: </b>Cyber physical systems emerge when physical systems are integrated with communication networks. In particular, communication networks facilitate dissemination of data among components of physical systems to meet key requirements, such as efficiency and reliability, in achieving an objective. In this dissertation, we consider one of the most important cyber physical systems: the smart grid. The North American Electric Reliability Corporation (NERC) envisions a smart grid that aggressively explores advance communication network solutions to facilitate real-time monitoring and dynamic control of the bulk electric power system. At the distribution level, the smart grid integrates renewable generation and energy storage mechanisms to improve reliability of the grid. Furthermore, dynamic pricing and demand management provide customers an avenue to interact with the power system to determine electricity usage that satisfies their lifestyle. At the transmission level, efficient communication and a highly automated architecture provide visibility in the power system; hence, faults are mitigated faster than they can propagate. However, higher levels of reliability and efficiency rely on the supporting physical communication infrastructure and the network technologies employed. Conventionally, the topology of the communication network tends to be identical to that of the power network. In this dissertation, however, we employ a Demand Response (DR) application to illustrate that a topology that may be ideal for the power network may not necessarily be ideal for the communication network. To develop this illustration, we realize that communication network issues, such as congestion, are addressed by protocols, middle-ware, and software mechanisms. Additionally, a network whose physical topology is designed to avoid congestion realizes an even higher level of performance. For this reason, characterizing the communication infrastructure of smart grids provides mechanisms to improve performance while minimizing cost. Most recently, algebraic connectivity has been used in the ongoing research effort characterizing the robustness of networks to failures and attacks. Therefore, we first derive analytical methods for increasing algebraic connectivity and validate these methods numerically. Secondly, we investigate impact on the topology and traffic characteristics as algebraic connectivity is increased. Finally, we construct a DR application to demonstrate how concepts from graph theory can dramatically improve the performance of a communication network. With a hybrid simulation of both power and communication network, we illustrate that a topology which may be ideal for the power network may not necessarily be ideal for the communication network. To date, utility companies are embracing network technologies such as Multiprotocol Label Switching (MPLS) because of the available support for legacy devices, traffic engineering, and virtual private networks (VPNs) which are essential to the functioning of the smart grid. Furthermore, this particular network technology meets the requirement of non-routability as stipulated by NERC, but these benefits are costly for the infrastructure that supports the full MPLS specification. More importantly, with MPLS routing and other switching technologies, innovation is restricted to the features provided by the equipment. In particular, no practical method exists for utility consultants or researchers to test new ideas, such as alternatives to IP or MPLS, on a realistic scale in order to obtain the experience and confidence necessary for real-world deployments. As a result, novel ideas remain untested. On the contrary, OpenFlow, which has gained support from network providers such as Microsoft and Google and equipment vendors such as NEC and Cisco, provides the programmability and flexibility necessary to enable innovation in next-generation communication architectures for the smart grid. This level of flexibility allows OpenFlow to provide all features of MPLS and allows OpenFlow devices to co-exist with existing MPLS devices. Therefore, in this dissertation we explore a low-cost OpenFlow Software Defined Networking solution and compare its performance to that of MPLS. In summary, we develop methods for designing robust networks and evaluate software defined networking for communication and control in cyber physical systems where the smart grid is the system under consideration.
     2848</li>
     2849<br>
     2850
     2851
     2852
     2853<li>
     2854<b>Sydney, Ali and Ochs, David S. and Scoglio, Caterina and Gruenbacher, Don and Miller, Ruth</b>
     2855, &quot;Using GENI for experimental evaluation of Software Defined Networking in smart grids.&quot;
     2856Computer Networks,
     28572014.
     2858doi:10.1016/j.bjp.2013.12.021.
     2859<a href="http://dx.doi.org/10.1016/j.bjp.2013.12.021">http://dx.doi.org/10.1016/j.bjp.2013.12.021</a>
     2860<br><br><b>Abstract: </b>The North American Electric Reliability Corporation (NERC) envisions a smart grid that aggressively explores advance communication network solutions to facilitate real-time monitoring and dynamic control of the bulk electric power system. At the distribution level, the smart grid integrates renewable generation and energy storage mechanisms to improve the reliability of the grid. Furthermore, dynamic pricing and demand management provide customers an avenue to interact with the power system to determine the electricity usage that best satisfies their lifestyle. At the transmission level, efficient communication and a highly automated architecture provide visibility in the power system and as a result, faults are mitigated faster than they can propagate. However, such higher levels of reliability and efficiency rest on the supporting communication infrastructure. To date, utility companies are moving towards Multiprotocol Label Switching (MPLS) because it supports traffic engineering and virtual private networks (VPNs). Furthermore, it provides Quality of Service (QoS) guarantees and fail-over mechanisms in addition to meeting the requirement of non-routability as stipulated by NERC. However, these benefits come at a cost for the infrastructure that supports the full MPLS specification. With this realization and given a two week implementation and deployment window in GENI, we explore the modularity and flexibility provided by the low cost OpenFlow Software Defined Networking (SDN) solution. In particular, we use OpenFlow to provide (1) automatic fail-over mechanisms, (2) a load balancing, and (3) Quality of Service guarantees: all essential mechanisms for smart grid networks.
     2861</li>
     2862<br>
     2863
     2864
     2865
     2866<li>
     2867<b>Tarui, Toshiaki and Kanada, Yasusi and Hayashi, Michiaki and Nakao, Akihiro</b>
     2868, &quot;Federating heterogeneous network virtualization platforms by slice exchange point.&quot;
     2869Integrated Network Management (IM), 2015 IFIP/IEEE International Symposium on, IEEE,
     28702015.
     2871doi:10.1109/inm.2015.7140366.
     2872<a href="http://dx.doi.org/10.1109/inm.2015.7140366">http://dx.doi.org/10.1109/inm.2015.7140366</a>
     2873<br><br><b>Abstract: </b>An architecture called the slice-exchange-point (SEP) has been designed for federating heterogeneous net-work-virtualization platforms by creating and managing slices (virtual networks). SEP enables whole inter-domain resources to be managed by the network manager of any single domain. Slice-operation commands are propagated to other domains through SEP by using a common API. SEP introduces the following four features: infrastructure neutrality, single interface federation, abstract and clean federation, and extensibility of capabilities. SEP's functions to achieve these features are discussed. SEP was partially implemented on two VNode domains and one ProtoGENI domain and was verified to function effectively.
     2874</li>
     2875<br>
     2876
     2877
     2878
     2879<li>
     2880<b>Teerapittayanon, Surat and Fouli, Kerim and M&#x65;&#x0301;dard, Muriel and Montpetit, Marie-Jos&#x65;&#x0301; and Shi, Xiaomeng and Seskar, Ivan and Gosain, Abhimanyu</b>
     2881, &quot;Network Coding as a WiMAX Link Reliability Mechanism.&quot;
     2882Multiple Access Communications, Springer Berlin Heidelberg,
     28832012.
     2884doi:10.1007/978-3-642-34976-8&#x005F;1.
     2885<a href="http://dx.doi.org/10.1007/978-3-642-34976-8&#x005F;1">http://dx.doi.org/10.1007/978-3-642-34976-8&#x005F;1</a>
     2886<br><br><b>Abstract: </b>We design and implement a network-coding-enabled relia- bility architecture for next generation wireless networks. Our network coding (NC) architecture uses a flexible thread-based design, with each encoder-decoder instance applying systematic intra-session random lin- ear network coding as a packet erasure code at the IP layer. Using GENI WiMAX platforms, a series of point-to-point transmission experiments were conducted to compare the performance of the NC architecture to that of the Automatic Repeated reQuest (ARQ) and Hybrid ARQ (HARQ) mechanisms. In our scenarios, the proposed architecture is able to decrease packet loss from around 11-32&#x0025; to nearly 0&#x0025;; compared to HARQ and joint HARQ/ARQ mechanisms, the NC architecture offers up to 5.9 times gain in throughput and 5.5 times reduction in end-to- end file transfer delay. By establishing NC as a potential substitute for HARQ/ARQ, our experiments offer important insights into cross-layer designs of next generation wireless networks.
     2887</li>
     2888<br>
     2889
     2890
     2891
     2892<li>
     2893<b>Thomas, Charles and Sommers, Joel and Barford, Paul and Kim, Dongchan and Das, Ananya and Segebre, Roberto and Crovella, Mark</b>
     2894, &quot;A Passive Measurement System for Network Testbeds.&quot;
     2895Testbeds and Research Infrastructure. Development of Networks and Communities, Springer Berlin Heidelberg,
     28962012.
     2897doi:10.1007/978-3-642-35576-9&#x005F;14.
     2898<a href="http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;14">http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;14</a>
     2899<br><br><b>Abstract: </b>The ability to capture and process packet-level data is of intrinsic importance in network testbeds that offer broad experimental capabilities to researchers. In this paper we describe the design and implementation of a passive measurement system for network testbeds called GIMS. The system enables users to specify and centrally manage packet capture on a set of dedicated measurement nodes deployed on links in a distributed testbed. The first component of GIMS is a scalable experiment management system that coordinates multi-tenant access to measurement nodes through a web-based user interface. The second component of GIMS is a node management system that enables (i) local processing on packets (e.g., flow aggregation and sampling), (ii) meta-data to be added to captured packets (e.g., timestamps), (iii) packet anonymization per local security policy, and (iv) flexible data storage including transfer to remote archives. We demonstrate the capabilities of GIMS through a set of micro-benchmarks that specifically highlight the performance of the node management system deployed on a commodity workstation. Our implementations are openly available to the community and our development efforts are on-going.
     2900</li>
     2901<br>
     2902
     2903
     2904
     2905<li>
     2906<b>Tiako, Pierre F.</b>
     2907, &quot;Perspectives of delegation in team-based distributed software development over the GENI infrastructure (NIER track).&quot;
     2908Proceedings of the 33rd International Conference on Software Engineering, Waikiki, Honolulu, HI, USA, ACM, New York, NY, USA,
     29092011.
     2910doi:10.1145/1985793.1985905.
     2911<a href="http://dx.doi.org/10.1145/1985793.1985905">http://dx.doi.org/10.1145/1985793.1985905</a>
     2912<br><br><b>Abstract: </b>Team-based distributed software development (TBDSD) is one of the single biggest challenges facing software companies. The need to manage development efforts and resources in different locations increase the complexity and cost of modern day software development. Current software development environments do not provide suitable support to delegate task among teams with appropriate directives. TBDSD is also limited to the current internet capabilities. One of the resulting problems is the difficulty to delegate and control tasks assigned among remote teams. This paper proposes (1) a new framework for delegation in TBDSD, and (2) perspectives for deploying Process-centered Software Engineering Environments (PSEE) over the Global Environment for Network Innovations (GENI) infrastructure. GENI, the 'future Internet' that is taking shape in prototypes across the US, will allow, in the context of our study, to securely access and share software artifacts, resources, and tools as never before seen over the current Internet.
     2913</li>
     2914<br>
     2915
     2916
     2917
     2918<li>
     2919<b>Toseef, Umar and Pentikousis, Kostas</b>
     2920, &quot;Authentication and Authorization in FELIX.&quot;
     29212015 IEEE/ACM 8th International Conference on Utility and Cloud Computing (UCC), IEEE,
     29222015.
     2923doi:10.1109/ucc.2015.98.
     2924<a href="http://dx.doi.org/10.1109/ucc.2015.98">http://dx.doi.org/10.1109/ucc.2015.98</a>
     2925<br><br><b>Abstract: </b>FELIX, the EU-Japan jointly-funded project, establishes a software defined networking (SDN) experimental facility which spans two continents and several administrative domains via dynamic transit network connections. The FELIX architectural blueprint provides an excellent example where key topics such as policy-based software-defined infrastructure instantiation is supported by resource orchestrators which manage multi-domain distributed compute and network resources including on-demand provisioning of transit network resources. In this context, FELIX implements a modern approach for authentication and authorization in SDN experimental facilities which enables fine-grained control and avoids single points of failure. This paper details the underlying mechanisms for user and transit network resource authentication and authorization in FELIX.
     2926</li>
     2927<br>
     2928
     2929
     2930
     2931<li>
     2932<b>Tredger, S. and Zhuang, Yanyan and Matthews, C. and Short-Gershman, J. and Coady, Y. and McGeer, R.</b>
     2933, &quot;Building Green Systems with Green Students: An Educational Experiment with GENI Infrastructure.&quot;
     2934Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     29352013.
     2936doi:10.1109/gree.2013.15.
     2937<a href="http://dx.doi.org/10.1109/gree.2013.15">http://dx.doi.org/10.1109/gree.2013.15</a>
     2938<br><br><b>Abstract: </b>Experimentation in system-oriented courses is often challenging, due to the raw and complex nature of the underlying infrastructure. In this work, we present our findings in teaching cloud computing to upper-level and graduate level students with GENI testbeds that are in use by the distributed systems community. The possibility of giving students practical and relevant experience was explored in the context of new course assignment objectives. Furthermore, students were able to explore systems concepts using GENI testbeds, and contribute to a collaborative class wide project with medium scale computation using satellite data. Our proposed set of experiments and course project provide a basis for an evaluation of the tradeoffs of teaching cloud and distributed systems. However, the software engineering challenges in these environments proved to be daunting. The amount of installation, configuration, and maintenance of their experiments was more than what students anticipated. The challenges the students faced drove them towards more traditional local development than attempting to work on the testbeds we presented. We hope that our findings provide insight into some of the possibilities to consider when preparing the next generation of computer scientists to engage with software practices and paradigms that are already fundamental in today's highly distributed systems.
     2939</li>
     2940<br>
     2941
     2942
     2943
     2944<li>
     2945<b>Tredger, Stephen</b>
     2946, &quot;SageFS: The Location Aware Wide Area Distributed Filesystem (Master's thesis).&quot;
     2947
     29482014.
     2949
     2950<a href="http://dspace.library.uvic.ca/bitstream/handle/1828/5824/Tredger&#x005F;Stephen&#x005F;MSc&#x005F;2014.pdf?sequence=3&#x0026;&#x0023;38;isAllowed=y">http://dspace.library.uvic.ca/bitstream/handle/1828/5824/Tredger&#x005F;Stephen&#x005F;MSc&#x005F;2014.pdf?sequence=3&#x0026;&#x0023;38;isAllowed=y</a>
     2951<br><br><b>Abstract: </b>Modern distributed applications often have to make a choice about how to maintain data within the system. Distributed storage systems are often self- contained in a single cluster or are a black box as data placement is unknown by an application. Using wide area distributed storage either means using multiple APIs or loss of control of data placement. This work introduces Sage, a distributed filesystem that aggregates multiple backends under a common API. It also gives applications the ability to decide where file data is stored in the aggregation. By leveraging Sage, users can create applications using multiple distributed backends with the same API, and still decide where to physically store any given file. Sage uses a layered design where API calls are translated into the appropriate set of backend calls then sent to the correct physical backend. This way Sage can hold many backends at once making them appear as the same filesystem. The performance overhead of using Sage is shown to be minimal over directly using the backend stores, and Sage is also shown to scale with respect to backends used. A case study shows file placement in action and how applications can take advantage of the feature.
     2952</li>
     2953<br>
     2954
     2955
     2956
     2957<li>
     2958<b>Tsai, Pang-Wei and wen Cheng, Pei and Yang, Chu-Sing and Luo, Mon-Yen</b>
     2959, &quot;Supporting Extensions of VLAN-tagged traffic across OpenFlow Networks.&quot;
     29602013 Proceedings Second GENI Research and Educational Experiment Workshop, Salt Lake City, UT, IEEE,
     29612013.
     2962doi:10.1109/GREE.2013.20.
     2963<a href="http://dx.doi.org/10.1109/GREE.2013.20">http://dx.doi.org/10.1109/GREE.2013.20</a>
     2964
     2965</li>
     2966<br>
     2967
     2968
     2969
     2970<li>
     2971<b>Tuncer, Hasan and Nozaki, Yoshihiro and Shenoy, Nirmala</b>
     2972, &quot;Virtual Mobility Domains - A Mobility Architecture for the Future Internet.&quot;
     2973IEEE International Conference on Commnunications (IEE ICC 2012) Symposium on Next-Generation Networking,
     29742012.
     2975doi:10.1109/ICC.2012.6363872.
     2976<a href="ftp://lesc.det.unifi.it/pub/LenLar/proceedings/2012/ICC2012/symposia/papers/virtual&#x005F;mobility&#x005F;domains&#x005F;-&#x005F;a&#x005F;mobility&#x005F;architecture&#x005F;for&#x005F;the&#x005F;\\_.pdf">ftp://lesc.det.unifi.it/pub/LenLar/proceedings/2012/ICC2012/symposia/papers/virtual&#x005F;mobility&#x005F;domains&#x005F;-&#x005F;a&#x005F;mobility&#x005F;architecture&#x005F;for&#x005F;the&#x005F;\\_.pdf</a>
     2977<br><br><b>Abstract: </b>This paper presents a novel mobility architecture called Virtual Mobility Domains that is designed to work with the Floating Cloud Tiered Internetworking model. Virtual Mobility Domains supports both inter Autonomous System (macro) and intra Autonomous System (micro) mobility by leveraging a tiered addressing, a network cloud concept, and a unique packet forwarding scheme introduced by the Floating Cloud Tiered Internetworking model. The proposed mobility architecture is distinct from others by not using IP addressing and classic routing protocols, and deploying user-centric overlapping mobility domains. The comparative simulation study of Virtual Mobility Domains against Mobile IPv6, Hierarchical Mobile IPv6, and Proxy Mobile IPv6 using OPNET shows that Virtual Mobility Domains brings lower latency, lesser signaling overhead, and fewer packets loss during handoffs, specially during inter Autonomous System roaming. The results highlight the potential for a seamless mobility management.
     2978</li>
     2979<br>
     2980
     2981
     2982
     2983<li>
    842984<b>Turner, Jonathan S.</b>
    852985, &quot;A proposed architecture for the GENI backbone platform.&quot;
     
    942994
    952995
    96 <br>
    97 <a id="full-2007"><H2>GENI Publications for 2007</H2></a>
    98 
    99 
    100 <li>
    101 <b>Feamster, Nick and Gao, Lixin and Rexford, Jennifer</b>
    102 , &quot;How to lease the internet in your spare time.&quot;
    103 SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
    104 2007.
    105 doi:10.1145/1198255.1198265.
    106 <a href="http://doi.acm.org/10.1145/1198255.1198265">http://doi.acm.org/10.1145/1198255.1198265</a>
    107 <br><br><b>Abstract: </b>Today's Internet Service Providers (ISPs) serve two roles: managing their network infrastructure and providing (arguably limited) services to end users. We argue that coupling these roles impedes the deployment of new protocols and architectures, and that the future Internet should support two separate entities: infrastructure providers (who manage the physical infrastructure) and service providers (who deploy network protocols and offer end-to-end services). We present a high-level design for Cabo, an architecture that enables this separation; we also describe challenges associated with realizing this architecture
    108 </li>
    109 <br>
    110 
    111 
    112 
    1132996<li>
    1142997<b>Turner, Jonathan S. and Crowley, Patrick and DeHart, John and Freestone, Amy and Heller, Brandon and Kuhns, Fred and Kumar, Sailesh and Lockwood, John and Lu, Jing and Wilson, Michael and Wiseman, Charles and Zar, David</b>
     
    1373020
    1383021
    139 <br>
    140 <a id="full-2008"><H2>GENI Publications for 2008</H2></a>
    141 
    142 
    143 <li>
    144 <b>Herron, Jon-Paul</b>
    145 , &quot;GENI Meta-Operations Center.&quot;
    146 2008 IEEE Fourth International Conference on eScience, Indianapolis, IN, USA, IEEE,
    147 2008.
    148 doi:10.1109/eScience.2008.103.
    149 <a href="http://dx.doi.org/10.1109/eScience.2008.103">http://dx.doi.org/10.1109/eScience.2008.103</a>
    150 <br><br><b>Abstract: </b>NSF's GENI program represents an opportunity to build the kind of programmable, virtualized testbed scientists exploring the future of networking will need to support their research. As with any other scientific instrument, it will be crucial that the GENI infrastructure offer repeatable, consistent results to the researchers using it.The GENI Meta-Operations Center, operated by the Global Research NOC at Indiana University, will develop the software, protocols, and processes needed to ensure the repeatability, consistency, and efficiency of GENI.
    151 </li>
    152 <br>
    153 
    154 
    155 
    156 <li>
    157 <b>Mahindra, R. and Bhanage, G. D. and Hadjichristofi, G. and Seskar, I. and Raychaudhuri, D. and Zhang, Y. Y.</b>
    158 , &quot;Space Versus Time Separation for Wireless Virtualization on an Indoor Grid.&quot;
    159 Next Generation Internet Networks, 2008. NGI 2008, IEEE,
    160 2008.
    161 doi:10.1109/NGI.2008.36.
    162 <a href="http://dx.doi.org/10.1109/NGI.2008.36">http://dx.doi.org/10.1109/NGI.2008.36</a>
    163 <br><br><b>Abstract: </b>The decreasing cost of wireless hardware and ever increasing number of wireless testbeds has led to a shift in the protocol evaluation paradigm from simulations towards emulation. In addition, with a large number of users demanding experimental resources and lack of space and time for deploying more hardware, fair resource sharing among independent co-existing experiments is important. We study the proposed approaches to wireless virtualization with a focus on schemes conserving wireless channels rather than nodes. Our detailed comparison reveals that while experiments sharing a channel by space separation achieve better efficiency than those relying on time separation of a channel, the isolation between experiments in both cases is comparable. We propose and implement a policy manager to alleviate the isolation problem and suggest scenarios in which either of the schemes would provide a suitable virtualization solution.
    164 </li>
    165 <br>
    166 
    167 
    168 
    169 <li>
    170 <b>Mahindra, R. and Bhanage, G. and Hadjichristofi, G. and Ganu, S. and Kamat, P. and Seskar, I. and Raychaudhuri, D.</b>
    171 , &quot;Integration of heterogeneous networking testbeds.&quot;
    172 Proceedings of the 4th International Conference on Testbeds and research infrastructures for the development of networks &#x0026; communities, Innsbruck, Austria, ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), ICST, Brussels, Belgium, Belgium,
    173 2008.
    174 
    175 <a href="http://portal.acm.org/citation.cfm?id=1390609">http://portal.acm.org/citation.cfm?id=1390609</a>
    176 <br><br><b>Abstract: </b>As networking research expands into new frontiers, the research community has felt a need for a heterogeneous networking research infrastructure to experiment with the interaction and integration of different types of networks, and to test the performance of various networking protocols in realistic environments. This requirement has led to the Global Environment for Network Innovations (GENI) initiative to create a global infrastructure for conducting networking experiments across diverse substrates such as wired (local and wide-area), wireless, sensor and cellular networks. In this paper, we discuss and present two models for building such an experimental infrastructure. The first model enables a wired testbed to link with wireless edge nodes during an experiment, whereas the second model enables a wireless testbed to link to wired testbeds. Proof-of-concept experiments are also presented reinforcing the usefulness of the models in terms of facilitating experiments over the integrated heterogeneous infrastructure.
    177 </li>
    178 <br>
    179 
    180 
    181 
    182 <li>
    183 <b>McKeown, Nick and Anderson, Tom and Balakrishnan, Hari and Parulkar, Guru and Peterson, Larry and Rexford, Jennifer and Shenker, Scott and Turner, Jonathan</b>
    184 , &quot;OpenFlow: enabling innovation in campus networks.&quot;
    185 SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
    186 2008.
    187 doi:10.1145/1355734.1355746.
    188 <a href="http://doi.acm.org/10.1145/1355734.1355746">http://doi.acm.org/10.1145/1355734.1355746</a>
    189 <br><br><b>Abstract: </b>This whitepaper proposes OpenFlow: a way for researchers to run experimental protocols in the networks they use every day. OpenFlow is based on an Ethernet switch, with an internal flow-table, and a standardized interface to add and remove flow entries. Our goal is to encourage networking vendors to add OpenFlow to their switch products for deployment in college campus backbones and wiring closets. We believe that OpenFlow is a pragmatic compromise: on one hand, it allows researchers to run experiments on heterogeneous switches in a uniform way at line-rate and with high port-density; while on the other hand, vendors do not need to expose the internal workings of their switches. In addition to allowing researchers to evaluate their ideas in real-world traffic settings, OpenFlow could serve as a useful campus component in proposed large-scale testbeds like GENI. Two buildings at Stanford University will soon run OpenFlow networks, using commercial Ethernet switches and routers. We will work to encourage deployment at other schools; and We encourage you to consider deploying OpenFlow in your university network too.
    190 </li>
    191 <br>
    192 
    193 
    194 
    195 <li>
    196 <b>Scoglio, Caterina M. and Sydney, Ali and Youssef, Mina and Schumm, Phillip and Kooij, Robert E.</b>
    197 , &quot;Elasticity and Viral Conductance: Unveiling Robustness in Complex Networks through Topological Characteristics.&quot;
    198 CoRR,
    199 2008.
    200 
    201 <a href="http://arxiv-web3.library.cornell.edu/abs/0811.3272v3">http://arxiv-web3.library.cornell.edu/abs/0811.3272v3</a>
    202 <br><br><b>Abstract: </b>With increasingly ambitious initiatives such as GENI and FIND that seek to design the future Internet, it becomes imperative to define the characteristics of robust topologies, and build future networks optimized for robustness. This paper investigates the characteristics of network topologies that maintain a high level of throughput in spite of multiple attacks. To this end, we select network topologies belonging to the main network models and some real world networks. We consider three types of attacks: removal of random nodes, high degree nodes, and high betweenness nodes. We use elasticity as our robustness measure and, through our analysis, illustrate that different topologies can have different degrees of robustness. In particular, elasticity can fall as low as 0.8&#x0025; of the upper bound based on the attack employed. This result substantiates the need for optimized network topology design. Furthermore, we implement a tradeoff function that combines elasticity under the three attack strategies and considers the cost of the network. Our extensive simulations show that, for a given network density, regular and semi-regular topologies can have higher degrees of robustness than heterogeneous topologies, and that link redundancy is a sufficient but not necessary condition for robustness.
    203 </li>
    204 <br>
    205 
    206 
    207 
    208 <br>
    209 <a id="full-2009"><H2>GENI Publications for 2009</H2></a>
    210 
    211 
    212 <li>
    213 <b>Albrecht, Jeannie R.</b>
    214 , &quot;Bringing big systems to small schools: distributed systems for undergraduates.&quot;
    215 SIGCSE Bull., ACM, New York, NY, USA,
    216 2009.
    217 doi:10.1145/1539024.1508903.
    218 <a href="http://dx.doi.org/10.1145/1539024.1508903">http://dx.doi.org/10.1145/1539024.1508903</a>
    219 <br><br><b>Abstract: </b>Distributed applications have become a core component of the Internet's infrastructure. However, many undergraduate curriculums, especially at small colleges, do not offer courses that focus on the design and implementation of distributed systems. The courses that are offered address the theoretical aspects of system design, but often fail to provide students with the opportunity to develop and evaluate distributed applications in real-world environments. As a result, undergraduate students are not as prepared as they should be for graduate study or careers in industry. This paper describes an undergraduate course in Distributed Systems that not only studies the key design principles of distributed systems, but also has a unique emphasis on giving students hands-on access to distributed systems through the use of shared computing testbeds, such as PlanetLab and GENI, and open-source technologies, such as Xen and Hadoop. Using these platforms, students can perform large-scale, distributed experimentation even at small colleges.
    220 </li>
    221 <br>
    222 
    223 
    224 
    225 <li>
    226 <b>Baldine, I.</b>
    227 , &quot;Unique optical networking facilities and cross-layer networking.&quot;
    228 Summer Topical Meeting, 2009. LEOSST '09. IEEE/LEOS,
    229 2009.
    230 doi:10.1109/LEOSST.2009.5226210.
    231 <a href="http://dx.doi.org/10.1109/LEOSST.2009.5226210">http://dx.doi.org/10.1109/LEOSST.2009.5226210</a>
    232 
    233 </li>
    234 <br>
    235 
    236 
    237 
    238 <li>
    239 <b>Baldine, Ilia and Xin, Yufeng and Evans, Daniel and Heerman, Chris and Chase, Jeff and Marupadi, Varun and Yumerefendi, Aydan</b>
    240 , &quot;The missing link: Putting the network in networked cloud computing.&quot;
    241 in ICVCI09: International Conference on the Virtual Computing Initiative,
    242 2009.
    243 
    244 
    245 
    246 </li>
    247 <br>
    248 
    249 
    250 
    251 <li>
    252 <b>Elliott, Chip and Falk, Aaron</b>
    253 , &quot;An update on the GENI project.&quot;
    254 SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
    255 2009.
    256 doi:10.1145/1568613.1568620.
    257 <a href="http://dx.doi.org/10.1145/1568613.1568620">http://dx.doi.org/10.1145/1568613.1568620</a>
    258 <br><br><b>Abstract: </b>Environment for Network Innovations. Early prototypes of GENI are starting to come online as an end-to-end system and network researchers are invited to participate by engaging in the design process or using GENI to conduct experiments.
    259 </li>
    260 <br>
    261 
    262 
    263 
    264 <li>
    265 <b>Kim, Dae Y. and Mathy, Laurent and Campanella, Mauro and Summerhill, Rick and Williams, James and Shimojo, Shinji and Kitamura, Yasuichi and Otsuki, Hideaki</b>
    266 , &quot;Future Internet: Challenges in Virtualization and Federation.&quot;
    267 2009 Fifth Advanced International Conference on Telecommunications, Venice/Mestre, Italy, IEEE,
    268 2009.
    269 doi:10.1109/AICT.2009.8.
    270 <a href="http://dx.doi.org/10.1109/AICT.2009.8">http://dx.doi.org/10.1109/AICT.2009.8</a>
    271 <br><br><b>Abstract: </b>Future Internet is a clean-slate research activity in the quest of new networking technologies to overcome the limits of the current Internet. In its experimental research, virtualization and federation are emerging as essential features, especially in the construction and operation of the testbeds. Moreover, they are believed to sustain as the fundamental features of the Future Internet itself. Visions and experiences on virtualization and federation are given by leading experts from US, EU, and Asia.
    272 </li>
    273 <br>
    274 
    275 
    276 
    277 <br>
    278 <a id="full-2010"><H2>GENI Publications for 2010</H2></a>
    279 
    280 
    281 <li>
    282 <b>Albrecht, J. and Huang, D. Y.</b>
    283 , &quot;Managing distributed applications using Gush.&quot;
    284 Proceedings of the ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities, Testbed Practices Session (TridentCom),
    285 2010.
    286 doi:10.1007/978-3-642-17851-1&#x005F;31.
    287 <a href="http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;31">http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;31</a>
    288 <br><br><b>Abstract: </b>Deploying and controlling experiments running on a distributed set of resources is a challenging task. Software developers often spend a significant amount of time dealing with the complexities associated with resource configuration and management in these environments. Experiment control systems are designed to automate the process, and to ultimately help developers cope with the common problems that arise during the design, implementation, and evaluation of distributed systems. However, many of the existing control systems were designed with specific computing environments in mind, and thus do not provide support for heterogeneous resources in different testbeds. In this paper, we explore the functionality of Gush, an experiment control system, and discuss how it supports execution on three of the four GENI control frameworks.
    289 </li>
    290 <br>
    291 
    292 
    293 
    294 <li>
    295 <b>Baldine, Ilia and Xin, Yufeng and Mandal, Anirban and Renci, Chris H. and Chase, Unc-Ch J. and Marupadi, Varun and Yumerefendi, Aydan and Irwin, David</b>
    296 , &quot;Networked cloud orchestration: A GENI perspective.&quot;
    297 2010 IEEE Globecom Workshops, Miami, FL, USA, IEEE,
    298 2010.
    299 doi:10.1109/GLOCOMW.2010.5700385.
    300 <a href="http://dx.doi.org/10.1109/GLOCOMW.2010.5700385">http://dx.doi.org/10.1109/GLOCOMW.2010.5700385</a>
    301 <br><br><b>Abstract: </b>This paper describes the experience of developing a system for creation of distributed linked configurations of heterogeneous resources (slices) in GENI. Our work leverages a number of unique architectural solutions (distributed architecture, declarative resource specifications, unique approach to slice instantiation) which is applicable to a wider set of problems related to autonomic co-scheduling and provisioning of heterogeneous networked resources. We discuss the architecture, the resource description mechanisms and some of the algorithms used to enable our system. We conclude with an analysis of a real experiment at allocating resources from multiple providers across a very wide geographic area (spanning Massachusetts, Illinois and North Carolina) to create a single private Layer 2 network connecting virtual machines on the campus of Duke University to a sensor testbed at University of Massachusetts, Amherst.
    302 </li>
    303 <br>
    304 
    305 
    306 
    307 <li>
    308 <b>Bhanage, G. and Daya, R. and Seskar, I. and Raychaudhuri, D.</b>
    309 , &quot;VNTS: A Virtual Network Traffic Shaper for Air Time Fairness in 802.16e Systems.&quot;
    310 Communications (ICC), 2010 IEEE International Conference on, IEEE,
    311 2010.
    312 doi:10.1109/ICC.2010.5502484.
    313 <a href="http://dx.doi.org/10.1109/ICC.2010.5502484">http://dx.doi.org/10.1109/ICC.2010.5502484</a>
    314 <br><br><b>Abstract: </b>The 802.16e standard for broadband wireless access mandates the presence of QoS classes, but does not specify guidelines for the scheduler implementation or mechanisms to ensure air time fairness. Our study demonstrates the feasibility of controlling downlink airtime fairness for slices while running above a proprietary WiMAX basestation (BS) scheduler. We design and implement a virtualized infrastructure that allows users to obtain at least an allocated percentage of BS resources in the presence of saturation and link degradation. Using Kernel virtual machines for creating slices and Click modular router for implementing the virtual network traffic shaping engine we show that it is possible to adaptively control slice usage for downlink traffic on a WiMAX Basestation. The fairness index and coupling coefficient show an improvement of up to 42&#x0025;, and 73&#x0025; with preliminary indoor walking mobility experiments. Outdoor vehicular measurements show an improvement of up to 27&#x0025;, and 70\\\\ with the fairness index and coupling coefficient respectively
    315 </li>
    316 <br>
    317 
    318 
    319 
    320 <li>
    321 <b>Bhanage, G. and Vete, D. and Seskar, I. and Raychaudhuri, D.</b>
    322 , &quot;SplitAP: Leveraging Wireless Network Virtualization for Flexible Sharing of WLANs.&quot;
    323 Global Telecommunications Conference (GLOBECOM 2010), 2010 IEEE, IEEE,
    324 2010.
    325 doi:10.1109/GLOCOM.2010.5684328.
    326 <a href="http://dx.doi.org/10.1109/GLOCOM.2010.5684328">http://dx.doi.org/10.1109/GLOCOM.2010.5684328</a>
    327 <br><br><b>Abstract: </b>Providing air-time guarantees across a group of clients forms a fundamental building block in sharing an access point (AP) across different virtual network providers. Though this problem has a relatively simple solution for downlink group scheduling through traffic engineering at the AP, solving this problem for uplink (UL) traffic presents a challenge for fair sharing of wireless hotspots. Among other issues, the mechanism for uplink traffic control has to scale across a large user base, and provide flexible operation irrespective of the client channel conditions and network loads. In this study, we propose the SplitAP architecture that address the problem of sharing uplink airtime across groups of users by extending the idea of network virtualization. Our architecture allows us to deploy different algorithms for enforcing UL airtime fairness across client groups. In this study, we will highlight the design features of the SplitAP architecture, and present results from evaluation on a prototype deployed with: (1) LPFC and (2) LPFC+, two algorithms for controlling UL group fairness. Performance comparisons on the ORBIT testbed show that the proposed algorithms are capable of providing group air-time fairness across wireless clients irrespective of the network volume, and traffic type. The algorithms show up to 40&#x0025; improvement with a modified Jain fairness index.
    328 </li>
    329 <br>
    330 
    331 
    332 
    333 <li>
    334 <b>Bhanage, Gautam and Seskar, Ivan and Mahindra, Rajesh and Raychaudhuri, Dipankar</b>
    335 , &quot;Virtual basestation: architecture for an open shared WiMAX framework.&quot;
    336 Proceedings of the second ACM SIGCOMM workshop on Virtualized infrastructure systems and architectures, New Delhi, India, ACM, New York, NY, USA,
    337 2010.
    338 doi:10.1145/1851399.1851401.
    339 <a href="http://dx.doi.org/10.1145/1851399.1851401">http://dx.doi.org/10.1145/1851399.1851401</a>
    340 <br><br><b>Abstract: </b>This paper presents the architecture and performance evaluation of a virtualized wide-area &#x34;&#x0308;G&#x20;&#x0308;cellular wireless network. Specifically, it addresses the challenges of virtualization of resources in a cellular base station to enable shared use by multiple independent slice users (experimenters or mobile virtual network operators), each with possibly distinct flow types and network layer protocols. The proposed virtual basestation architecture is based on an external substrate which uses a layer-2 switched datapath, and an arbitrated control path to the WiMAX basestation. The framework implements virtualization of base station's radio resources to achieve isolation between multiple virtual networks. An algorithm for weighted fair sharing among multiple slices based on an airtime fairness metric has been implemented for the first release. Preliminary experimental results from the virtual basestation prototype are given, demonstrating mobile network performance, isolation across slices with different flow types, and custom flow scheduling capabilities.
    341 </li>
    342 <br>
    343 
    344 
    345 
    346 <li>
    347 <b>Chowdhury and Boutaba, Raouf</b>
    348 , &quot;A survey of network virtualization.&quot;
    349 Computer Networks,
    350 2010.
    351 doi:http://dx.doi.org/10.1016/j.comnet.2009.10.017.
    352 <a href="http://www.sciencedirect.com/science/article/pii/S1389128609003387">http://www.sciencedirect.com/science/article/pii/S1389128609003387</a>
    353 <br><br><b>Abstract: </b>Due to the existence of multiple stakeholders with conflicting goals and policies, alterations to the existing Internet architecture are now limited to simple incremental updates; deployment of any new, radically different technology is next to impossible. To fend off this ossification, network virtualization has been propounded as a diversifying attribute of the future inter-networking paradigm. By introducing a plurality of heterogeneous network architectures cohabiting on a shared physical substrate, network virtualization promotes innovations and diversified applications. In this paper, we survey the existing technologies and a wide array of past and state-of-the-art projects on network virtualization followed by a discussion of major challenges in this area.
    354 </li>
    355 <br>
    356 
    357 
    358 
    359 <li>
    360 <b>Erazo, Miguel A. and Liu, Jason</b>
    361 , &quot;On enabling real-time large-scale network simulation in GENI: the PrimoGENI approach.&quot;
    362 Proceedings of the 3rd International ICST Conference on Simulation Tools and Techniques, Torremolinos, Malaga, Spain, ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), ICST, Brussels, Belgium, Belgium,
    363 2010.
    364 doi:10.4108/ICST.SIMUTOOLS2010.8636.
    365 <a href="http://dx.doi.org/10.4108/ICST.SIMUTOOLS2010.8636">http://dx.doi.org/10.4108/ICST.SIMUTOOLS2010.8636</a>
    366 <br><br><b>Abstract: </b>The Global Environment for Network Innovations (GENI) is a community-driven research and development effort to build a collaborative and exploratory network experimentation platform, a &#x76;&#x0308;irtual laboratory&#x20;&#x0308;for the design, implementation and evaluation of future Internets. In this paper, we present an overview of PrimoGENI, a GENI project with the goal of extending the GENI suite of interoperable infrastructure to allow network experiments at scale, involving physical, simulated and emulated network entities.
    367 </li>
    368 <br>
    369 
    370 
    371 
    372 <li>
    373 <b>Feamster, Nick and Nayak, Ankur and Kim, Hyojoon and Clark, Russell and Mundada, Yogesh and Ramachandran, Anirudh and bin Tariq, Mukarram</b>
    374 , &quot;Decoupling policy from configuration in campus and enterprise networks.&quot;
    375 2010 17th IEEE Workshop on Local &#x0026; Metropolitan Area Networks (LANMAN), Long Branch, NJ, USA, IEEE,
    376 2010.
    377 doi:10.1109/LANMAN.2010.5507162.
    378 <a href="http://dx.doi.org/10.1109/LANMAN.2010.5507162">http://dx.doi.org/10.1109/LANMAN.2010.5507162</a>
    379 <br><br><b>Abstract: </b>This paper surveys our ongoing work on the use of software-defined networking to simplify two acute policy problems in campus and enterprise network operations: access control and information flow control. We describe how the current coupling of high-level policy with low-level configuration makes these problems challenging today. We describe the specific policy problems faced by campus and enterprise network operators; illustrate our approach, which leverages recent trends in separating the network's ” control plane” from the data plane; and show how this approach can be applied to simplify these two enterprise network management tasks. We also describe our ongoing deployment efforts to build a campus network testbed where trial designs can be deployed and evaluated. We close with a summary of current and future research challenges for solving challenges within enterprise networks within the context of this new paradigm.
    380 </li>
    381 <br>
    382 
    383 
    384 
    385 <li>
    386 <b>Sharma, Navin and Gummeson, Jeremy and Irwin, David and Shenoy, Prashant</b>
    387 , &quot;Cloudy Computing: Leveraging Weather Forecasts in Energy Harvesting Sensor Systems.&quot;
    388 2010 7th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON), Boston, MA, USA, IEEE,
    389 2010.
    390 doi:10.1109/SECON.2010.5508260.
    391 <a href="http://dx.doi.org/10.1109/SECON.2010.5508260">http://dx.doi.org/10.1109/SECON.2010.5508260</a>
    392 <br><br><b>Abstract: </b>To sustain perpetual operation, systems that harvest environmental energy must carefully regulate their usage to satisfy their demand. Regulating energy usage is challenging if a system's demands are not elastic and its hardware components are not energy-proportional, since it cannot precisely scale its usage to match its supply. Instead, the system must choose when to satisfy its energy demands based on its current energy reserves and predictions of its future energy supply. In this paper, we explore the use of weather forecasts to improve a system's ability to satisfy demand by improving its predictions. We analyze weather forecast, observational, and energy harvesting data to formulate a model that translates a weather forecast to a wind or solar energy harvesting prediction, and quantify its accuracy. We evaluate our model for both energy sources in the context of two different energy harvesting sensor systems with inelastic demands: a sensor testbed that leases sensors to external users and a lexicographically fair sensor network that maintains steady node sensing rates. We show that using weather forecasts in both wind- and solar-powered sensor systems increases each system's ability to satisfy its demands compared with existing prediction strategies.
    393 </li>
    394 <br>
    395 
    396 
    397 
    398 <li>
    399 <b>Sridharan, Mukundan and Zeng, Wenjie and Leal, William and Ju, Xi and Ramanath, Rajiv and Zhang, Hongwei and Arora, Anish</b>
    400 , &quot;From Kansei to KanseiGenie: Architecture of Federated, Programmable Wireless Sensor Fabrics.&quot;
    401 Proceedings of the ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TridentCom),
    402 2010.
    403 
    404 
    405 <br><br><b>Abstract: </b>This paper deals with challenges in federating wireless sensing fabrics. Federations of this sort are currently being developed in next generation global end-to-end experimentation infrastructures, such as GENI, to support rapid prototyping and hi-fidelity validation of protocols and applications. On one hand, federation should support access to diverse (and potentially provider-specific) wireless sensor resources and, on the other, it should enable users to uniformly task these resources. Instead of more simple basing federation upon a standard description of resources, we propose an architecture where the ontology of resource description can vary across providers, and a mapping of user needs to resources is performed to achieve uniform tasking. We illustrate one realization of this architecture, in terms of our refactoring the Kansei testbed to become the KanseiGenie federated fabric manager, which has full support for programmability, sliceability, and federated experimentation over heterogeneous sensing fabrics.
    406 </li>
    407 <br>
    408 
    409 
    410 
    4113022<li>
    4123023<b>Valancius, Vytautas and Feamster, Nick and Rexford, Jennifer and Nakao, Akihiro</b>
     
    4363047
    4373048<li>
    438 <b>Yuen, Marco</b>
    439 , &quot;GENI in the Cloud (Master's Thesis).&quot;
    440 
    441 2010.
    442 
    443 <a href="http://s3.amazonaws.com/marcoy&#x005F;thesis/Thesis.pdf">http://s3.amazonaws.com/marcoy&#x005F;thesis/Thesis.pdf</a>
    444 <br><br><b>Abstract: </b>Computer networking researchers often have access to a few dierent network testbeds (Section 1.2) for their experiments. However, those testbeds are limited in resources; contentions for resources are prominent in those testbeds especially when conference deadline is looming. Moreover, services running on those testbeds are subject to seasonal and daily trac spikes from users all round the world. Hence, demand for resources at the testbeds are high. Some researchers can use other testbeds in conjunction with the ones they are using. Even though each of the testbeds may have dierent infrastructures, and characteristics, in the end, what the researchers receive in return is a set of computing resources, either virtual machines or physical machines. Essentially, those testbeds are providing a similar service, but researchers have to manage the credentials for accessing the testbeds manually, and they have to manually request resources from dierent testbeds in order to setup experiments that span across dierent testbeds. This thesis presents GENICloud, a project that enables the federation of testbeds with clouds. Computing and storage resources can be provisioned to researchers and services running on existing testbeds dynamically from an Eucalyptus cloud. As a part of the GENICloud project, the user proxy (Section 3.4) provides a less arduous method for testbeds administrators to federate with other testbeds; the same serviceiv also manages researchers credentials, so they do not have to acquire resources from each testbed individually. The user proxy provides a single interface for researchers to interact with dierent testbeds and clouds and manage their experiments. Furthermore, GENICloud demonstrates that there are, in fact, quite a few architectural similarities between dierent testbeds and even clouds
    445 </li>
    446 <br>
    447 
    448 
    449 
    450 <br>
    451 <a id="full-2011"><H2>GENI Publications for 2011</H2></a>
    452 
    453 
    454 <li>
    455 <b>Albrecht, Jeannie and Tuttle, Christopher and Braud, Ryan and Dao, Darren and Topilski, Nikolay and Snoeren, Alex C. and Vahdat, Amin</b>
    456 , &quot;Distributed application configuration, management, and visualization with plush.&quot;
    457 ACM Trans. Internet Technol., ACM, New York, NY, USA,
    458 2011.
    459 doi:10.1145/2049656.2049658.
    460 <a href="http://dx.doi.org/10.1145/2049656.2049658">http://dx.doi.org/10.1145/2049656.2049658</a>
    461 <br><br><b>Abstract: </b>Support for distributed application management in large-scale networked environments remains in its early stages. Although a number of solutions exist for subtasks of application deployment, monitoring, and maintenance in distributed environments, few tools provide a unified framework for application management. Many of the existing tools address the management needs of a single type of application or service that runs in a specific environment, and these tools are not adaptable enough to be used for other applications or platforms. To this end, we present the design and implementation of Plush, a fully configurable application management infrastructure designed to meet the general requirements of several different classes of distributed applications. Plush allows developers to specifically define the flow of control needed by their computations using application building blocks. Through an extensible resource management interface, Plush supports execution in a variety of environments, including both live deployment platforms and emulated clusters. Plush also uses relaxed synchronization primitives for improving fault tolerance and liveness in failure-prone environments. To gain an understanding of how Plush manages different classes of distributed applications, we take a closer look at specific applications and evaluate how Plush provides support for each.
    462 </li>
    463 <br>
    464 
    465 
    466 
    467 <li>
    468 <b>Angu, Pragatheeswaran and Ramamurthy, Byrav</b>
    469 , &quot;Experiences with dynamic circuit creation in a regional network testbed.&quot;
    470 2011 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Shanghai, China, IEEE,
    471 2011.
    472 doi:10.1109/infcomw.2011.5928801.
    473 <a href="http://dx.doi.org/10.1109/infcomw.2011.5928801">http://dx.doi.org/10.1109/infcomw.2011.5928801</a>
    474 <br><br><b>Abstract: </b>In this paper we share our experiences of enabling dynamic circuit creation in the GpENI network. GpENI is a network research testbed in the mid-west USA involving several educational institutions. University of Nebraska-Lincoln is involved in provisioning dynamic circuits across the GpENI network among its participating universities. We discuss several options investigated for deploying dynamic circuits over the GpENI network as well as our demonstration experiments at the GENI engineering conferences. UNL has also collaborated with ProtoGENI project of University of Utah and Mid-Atlantic Crossroads (MAX) facility of Washington DC to create inter-domain dynamic circuits.
    475 </li>
    476 <br>
    477 
    478 
    479 
    480 <li>
    481 <b>Bhanage, Gautam and Seskar, Ivan and Zhang, Yanyong and Raychaudhuri, Dipankar and Jain, Shweta</b>
    482 , &quot;Experimental Evaluation of OpenVZ from a Testbed Deployment Perspective.&quot;
    483 Testbeds and Research Infrastructures. Development of Networks and Communities, Springer Berlin Heidelberg,
    484 2011.
    485 doi:10.1007/978-3-642-17851-1&#x005F;7.
    486 <a href="http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;7">http://dx.doi.org/10.1007/978-3-642-17851-1&#x005F;7</a>
    487 <br><br><b>Abstract: </b>A scalable approach to building large scale experimentation testbeds involves multiplexing the system resources for better utilization. Virtualization provides a convenient means of sharing testbed resources among experimenters. The degree of programmability and isolation achieved with such a setup is largely dependent on the type of technology used for virtualization. We consider OpenVZ and User Mode Linux (UML) for virtualization of the ORBIT wireless testbed and evaluate their relative merit. Our results show that OpenVZ, an operating system level virtualization mechanism significantly outperforms UML in terms of system overheads and performance isolation. We discuss both qualitative and quantitative performance features which could serve as guidelines for selection of a virtualization scheme for similar testbeds.
    488 </li>
    489 <br>
    490 
    491 
    492 
    493 <li>
    494 <b>Calyam, P. and Sridharan, M. and Xu, Yingxiao and Zhu, Kunpeng and Berryman, A. and Patali, R. and Venkataraman, A.</b>
    495 , &quot;Enabling performance intelligence for application adaptation in the Future Internet.&quot;
    496 Communications and Networks, Journal of,
    497 2011.
    498 doi:10.1109/JCN.2011.6157475.
    499 <a href="http://dx.doi.org/10.1109/JCN.2011.6157475">http://dx.doi.org/10.1109/JCN.2011.6157475</a>
    500 <br><br><b>Abstract: </b>Today's Internet which provides communication channels with best-effort end-to-end performance is rapidly evolving into an autonomic global computing platform. Achieving autonomicity in the Future Internet will require a performance architecture that (a) allows users to request and own 'slices' of geographically-distributed host and network resources, (b) measures and monitors end-to-end host and network status, (c) enables analysis of the measurements within expert systems, and (d) provides performance intelligence in a timely manner for application adaptations to improve performance and scalability. We de- scribe the requirements and design of one such &#x46;&#x0308;uture Internet performance architecture&#x20;&#x0308;(FIPA), and present our reference implementation of FIPA called 'OnTimeMeasure.' OnTimeMeasure comprises of several measurement-related services that can interact with each other and with existing measurement frameworks to enable performance intelligence. We also explain our OnTimeMea- sure deployment in the global environment for network innovations (GENI) infrastructure collaborative research initiative to build a sliceable Future Internet. Further, we present an application- adaptation case study in GENI that uses OnTimeMeasure-enabled performance intelligence in the context of dynamic resource allocation within thin-client based virtual desktop clouds. We show how a virtual desktop cloud provider in the Future Internet can use the performance intelligence to increase cloud scalability, while simultaneously delivering satisfactory user quality-of-experience.
    501 </li>
    502 <br>
    503 
    504 
    505 
    506 <li>
    507 <b>Chen, Kang and Shen, Haiying</b>
    508 , &quot;Global optimization of file availability through replication for efficient file sharing in MANETs.&quot;
    509 Network Protocols (ICNP), 2011 19th IEEE International Conference on, Vancouver, AB, Canada, IEEE,
    510 2011.
    511 doi:10.1109/icnp.2011.6089056.
    512 <a href="http://dx.doi.org/10.1109/icnp.2011.6089056">http://dx.doi.org/10.1109/icnp.2011.6089056</a>
    513 <br><br><b>Abstract: </b>File sharing applications in mobile ad hoc networks (MANETs) have attracted more and more attention in recent years. The efficiency of file querying suffers from the distinctive properties of MANETs including node mobility and limited communication range and resource. An intuitive method to alleviate this problem is to create file replicas in the network. However, despite the efforts on file replication, no research has focused on the global optimal replica sharing with minimum average querying delay. Specifically, current file replication protocols in MANETs have two shortcomings. First, they lack a rule to allocate limited resource to different files in order to minimize the average querying delay. Second, they simply consider storage as resource for replicas, but neglect the fact that the file holders' frequency of meeting other nodes also plays an important role in determining file availability. A node having a higher meeting frequency with others provides higher availability to its files. In this paper, we introduce a new concept of resource for file replication, which considers both node storage and meeting frequency. We theoretically study the influence of resource allocation on the average querying delay and derive a resource allocation rule to minimize the average querying delay. We further propose a distributed file replication protocol that follows the rule. The trace-driven experiments on both the real-world GENI testbed and NS-2 show that our protocol can achieve shorter average querying delay at lower cost than current replication protocols, which justifies the correctness of our theoretical analysis and the effectiveness of the proposed protocol.
    514 </li>
    515 <br>
    516 
    517 
    518 
    519 <li>
    520 <b>Chen, Kang and Shen, Haiying and Zhang, Haibo</b>
    521 , &quot;Leveraging Social Networks for P2P Content-Based File Sharing in Mobile Ad Hoc Networks.&quot;
    522 2011 IEEE Eighth International Conference on Mobile Ad-Hoc and Sensor Systems, Valencia, Spain, IEEE,
    523 2011.
    524 doi:10.1109/MASS.2011.24.
    525 <a href="http://dx.doi.org/10.1109/MASS.2011.24">http://dx.doi.org/10.1109/MASS.2011.24</a>
    526 <br><br><b>Abstract: </b>Current P2P file sharing methods in mobile ad hoc networks (MANETs) can be classified into three groups: flooding-based, advertisement-based and social contact-based. The first two groups of methods can easily generate high overhead and low scalability, and the third group fails to consider the social interests (content) of mobile nodes, which otherwise can improve file searching efficiency. In this paper, we propose a P2P content-based file sharing system for MANETs. The system uses an interest extraction algorithm to derive a node's interests from its files for complex queries. For efficient file searching, it groups common-interest nodes that frequently meet with each other as communities. Further, it takes advantage of node mobility by designating stable nodes, which has frequent contact with community members, as community coordinators for intra-community searching, and highly-mobile nodes as community ambassadors for inter-community searching. An interest-oriented file searching scheme further enhances the file searching success rate. We first deployed our system on the real-world GENI Orbit testbed with a real trace and then conducted experiment on the ns2 simulator with both real trace and simulated disconnected and connected MANET scenario. The test results show that our system significantly lowers transmission cost and improves file searching success rate compared to current methods.
    527 </li>
    528 <br>
    529 
    530 
    531 
    532 <li>
    533 <b>Cherukuri, Ramkumar and Liu, Xuan and Bavier, Andy and Sterbenz, James P. G. and Medhi, Deep</b>
    534 , &quot;Network virtualization in GpENI: Framework, implementation &#x0026;amp; integration experience.&quot;
    535 12th IFIP/IEEE International Symposium on Integrated Network Management (IM 2011) and Workshops, Dublin, Ireland, IEEE,
    536 2011.
    537 doi:10.1109/INM.2011.5990568.
    538 <a href="http://dx.doi.org/10.1109/INM.2011.5990568">http://dx.doi.org/10.1109/INM.2011.5990568</a>
    539 <br><br><b>Abstract: </b>Great Plains Environment for Network Innovation (GpENI) is an international testbed for future Internet research. A key component of GpENI is programmable network virtualization (GpENI-VINI). The scope of this paper is to present the framework, implementation and integration experience with network virtualization in GpENI. In particular, this is described through our experience of implementing and integrating the XORP (eXtensible Open Router Platform) routing platform into GpENI-VINI. Preliminary results on measurements and validation are presented.
    540 </li>
    541 <br>
    542 
    543 
    544 
    545 <li>
    546 <b>Das, S. and Yiakoumis, Y. and Parulkar, G. and McKeown, N. and Singh, P. and Getachew, D. and Desai, P. D.</b>
    547 , &quot;Application-aware aggregation and traffic engineering in a converged packet-circuit network.&quot;
    548 Optical Fiber Communication Conference and Exposition (OFC/NFOEC), 2011 and the National Fiber Optic Engineers Conference, IEEE,
    549 2011.
    550 
    551 <a href="http://ieeexplore.ieee.org/xpls/abs&#x005F;all.jsp?arnumber=5875210">http://ieeexplore.ieee.org/xpls/abs&#x005F;all.jsp?arnumber=5875210</a>
    552 <br><br><b>Abstract: </b>We demonstrate a converged OpenFlow enabled packet-circuit network, where circuit flow properties (guarantee d bandwidth, low latency, low jitter, bandwidth-on-demand, fast recovery) provide differential treatment to dynamically aggregated packet flows for voice, video and web traffic.
    553 </li>
    554 <br>
    555 
    556 
    557 
    558 <li>
    559 <b>Femminella, Mauro and Francescangeli, Roberto and Reali, Gianluca and Lee, Jae W. and Schulzrinne, Henning</b>
    560 , &quot;An enabling platform for autonomic management of the future internet.&quot;
    561 IEEE Network,
    562 2011.
    563 doi:10.1109/MNET.2011.6085639.
    564 <a href="http://dx.doi.org/10.1109/MNET.2011.6085639">http://dx.doi.org/10.1109/MNET.2011.6085639</a>
    565 <br><br><b>Abstract: </b>This article shows an autonomic management solution based on the recently defined programmable node architecture NetServ. The article starts with a general description of the classical network management requirements and their adaptation to the expected network evolution. After a description of the major issues characterizing the management of the expected Future Internet, the main autonomic management paradigms, and some recently introduced autonomic service platforms, we show and demonstrate the effectiveness of the NetServ architecture. Born as a means to deploy and execute networked services at runtime over programmable routers, NetServ has proved to be a suitable environment for hosting an autonomic management architecture.
    566 </li>
    567 <br>
    568 
    569 
    570 
    571 <li>
    572 <b>Gangam, Sriharsha and Fahmy, Sonia</b>
    573 , &quot;Mitigating interference in a network measurement service.&quot;
    574 2011 IEEE Nineteenth IEEE International Workshop on Quality of Service, San Jose, CA, USA, IEEE,
    575 2011.
    576 doi:10.1109/IWQOS.2011.5931347.
    577 <a href="http://dx.doi.org/10.1109/IWQOS.2011.5931347">http://dx.doi.org/10.1109/IWQOS.2011.5931347</a>
    578 <br><br><b>Abstract: </b>Shared measurement services offer key advantages over conventional ad-hoc techniques for network monitoring. A measurement service may receive measurement requests concurrently from different applications and network administrators. These measurement requests are often served by injecting active network measurement traffic between two hosts. Two active measurements are said to interfere when the probe packets of one measurement tool are viewed as network traffic by the other. This may lead to faulty measurement readings. In this paper, we model the measurement interference problem, and show how to schedule measurement tasks to reduce interference and hence increase measurement accuracy. We propose twelve computationally tractable algorithms that decrease the total completion time (makespan) of measurement tasks, while avoiding interference. Our evaluation shows that the algorithm we refer to as Largest Area First, Busiest Node First - Earliest Interval Schedule (LAFBNF-EIS) has a mean makespan of about 5&#x0025; more than the theoretical lower bound over our set of measurement workloads.
    579 </li>
    580 <br>
    581 
    582 
    583 
    584 <li>
    585 <b>Ju, Xi and Zhang, Hongwei and Zeng, Wenjie and Sridharan, Mukundan and Li, Jing and Arora, Anish and Ramnath, Rajiv and Xin, Yufeng</b>
    586 , &quot;LENS: resource specification for wireless sensor network experimentation infrastructures.&quot;
    587 Proceedings of the 6th ACM international workshop on Wireless network testbeds, experimental evaluation and characterization, Las Vegas, Nevada, USA, ACM, New York, NY, USA,
    588 2011.
    589 doi:10.1145/2030718.2030727.
    590 <a href="http://dx.doi.org/10.1145/2030718.2030727">http://dx.doi.org/10.1145/2030718.2030727</a>
    591 <br><br><b>Abstract: </b>As a first step towards predictable, repeatable WSN experimentation, we propose the resource specification language LENS (a.k.a. Language for Embedded Networked Sensing) for WSN experimentation infrastructures. Using the Resource Description Framework (RDF) and the Web Ontology Language (OWL), LENS defines a semantic ontology for WSN resources; LENS enables explicit control and measurement of uncertainty factors, and it enables reasoning about the relationships between WSN resources. Focusing on basic concepts of WSNs, LENS supports resource specification in a wide range of WSN experimentation infrastructures, and it is extensible to support potentially unforeseen technologies. LENS is also compatible with specification languages for other network resources such as optical networks. As a part of the NSF GENI initiative, we have implemented LENS in the KanseiGenie control framework, and LENS has been actively used to support experimentation in the federated WSN infrastructure involving Kansei and NetEye. Enabling reasoning about uncertainty factors in experimentation, LENS is expected to serve as a basis for developing methodologies and tools for predictable, repeatable WSN experimentation.
    592 </li>
    593 <br>
    594 
    595 
    596 
    597 <li>
    598 <b>Kangarlou, A. and Xu, Dongyan and Kozat, U. C. and Padala, P. and Lantz, B. and Igarashi, K.</b>
    599 , &quot;In-network live snapshot service for recovering virtual infrastructures.&quot;
    600 Network, IEEE, IEEE,
    601 2011.
    602 doi:10.1109/mnet.2011.5958003.
    603 <a href="http://dx.doi.org/10.1109/mnet.2011.5958003">http://dx.doi.org/10.1109/mnet.2011.5958003</a>
    604 <br><br><b>Abstract: </b>Infrastructure as a Service (IaaS) has become an increasingly popular type of service for both private and public clouds. The virtual infrastructures that enable IaaS support multitenancy by multiplexing the computational resources of data centers and result in substantial reductions in operational costs. Since hardware and software failures occur on a routine basis in large-scale systems, it is imperative for cloud providers to offer various failure recovery options for distributed services hosted on such infrastructures. In this article we present GENI-VIOLIN, a new cloud capability that can checkpoint a stateful distributed service while incurring very low overhead. The unique aspect of GENI-VIOLIN compared to previous work is that GENI-VIOLIN exploits programmable OpenFlow switches to provide checkpointing services in the network, thereby requiring minimal changes to the end host virtualization framework. We have developed a prototype of GENI-VIOLIN using the GENI infrastructure, and have demonstrated GENI-VIOLIN's checkpoint and restore capability across multiple GENI sites.
    605 </li>
    606 <br>
    607 
    608 
    609 
    610 <li>
    611 <b>Katz-Bassett, Ethan and Choffnes, David R. and Cunha, &#x49;&#x0301;talo and Scott, Colin and Anderson, Thomas and Krishnamurthy, Arvind</b>
    612 , &quot;Machiavellian Routing: Improving Internet Availability with BGP Poisoning.&quot;
    613 Proceedings of the 10th ACM Workshop on Hot Topics in Networks, Cambridge, Massachusetts, ACM, New York, NY, USA,
    614 2011.
    615 doi:10.1145/2070562.2070573.
    616 <a href="http://dx.doi.org/10.1145/2070562.2070573">http://dx.doi.org/10.1145/2070562.2070573</a>
    617 <br><br><b>Abstract: </b>We propose a new approach to mitigate disruptions of Internet connectivity. The Internet was designed to always find a route if there is a policy-compliant path; however, in many cases, connectivity is disrupted despite the existence of an underlying valid path. The research community has done considerable work on this problem, much of it focused on short-term outages that occur during route convergence. There has been less progress on addressing avoidable long-lasting outages. Our measurements show that long-lasting events contribute significantly to overall unavailability. To address these long-term problems, we develop a system, Machiavellian routing, for automatic failure remediation, centered around the use of BGP poisoning. With poisoning, an edge network can cause other networks to send traffic to it via paths that avoid a problem in a particular transit ISP. We describe the key challenges to using poisoning to improve Internet connectivity, and we develop a set of techniques to use it predictably, accurately, and effectively.
    618 </li>
    619 <br>
    620 
    621 
    622 
    623 <li>
    624 <b>Kline, Donald and Quan, John</b>
    625 , &quot;Attribute description service for large-scale networks.&quot;
    626 Proceedings of the 2nd international conference on Human centered design, Orlando, FL, USA, Springer-Verlag, Berlin, Heidelberg,
    627 2011.
    628 doi:10.1007/978-3-642-21753-1&#x005F;58.
    629 <a href="http://portal.acm.org/citation.cfm?id=2021672.2021735">http://portal.acm.org/citation.cfm?id=2021672.2021735</a>
    630 <br><br><b>Abstract: </b>An analysis of requesting resources from large-scale networks reveals a fundamental challenge. As the network grows, more and more resources become available, and so finding resources that fit experimental test criteria becomes difficult and time consuming. For example, the National Science Foundation sponsors GENI--an experimental network with a goal to gain enough resources to model the Internet at scale. Currently, GENI contains relatively few contributed resources donated from businesses and academia, and so matching resources to tests is rather simple. However, experimenters plan to conduct network experiments that are very complex and difficult to accurately model by using the vast numbers of resources expected in GENI. When GENI reaches its final state, finding the right resources that fit experimental test criteria out of many thousands of donated resources may be as difficult as conducting the experiment itself. This dilemma underscores the importance of establishing an attribute description service that promotes a standardized language for all interactions between the end users and the large-scale network.
    631 </li>
    632 <br>
    633 
    634 
    635 
    636 <li>
    637 <b>Lee, Jae W. and Francescangeli, Roberto and Janak, Jan and Srinivasan, Suman and Baset, Salman A. and Schulzrinne, Henning and Despotovic, Zoran and Kellerer, Wolfgang</b>
    638 , &quot;NetServ: Active Networking 2.0.&quot;
    639 2011 IEEE International Conference on Communications Workshops (ICC), Kyoto, Japan, IEEE,
    640 2011.
    641 doi:10.1109/iccw.2011.5963554.
    642 <a href="http://dx.doi.org/10.1109/iccw.2011.5963554">http://dx.doi.org/10.1109/iccw.2011.5963554</a>
    643 <br><br><b>Abstract: </b>We present NetServ, a node architecture for deploying in-network services in the next generation Internet. NetServ-enabled network nodes provide a common execution environment, where network services implemented as modules can be dynamically installed and removed. We demonstrate three such modules. MicroCDN is a dynamic content distribution network (CDN) service which implements a content caching strategy specific to a content provider. The NAT Keep-alive module offloads the processing of keep-alive messages from SIP servers. The Media Relay module allows any NetServ node to act as a media relay, eliminating the need to manage standalone relay servers. NetServ aims to revive the Active Networking vision. It was too far ahead of its time a decade ago, but we believe its time has finally arrived.
    644 </li>
    645 <br>
    646 
    647 
    648 
    649 <li>
    650 <b>Li, Dawei and Hong, Xiaoyan</b>
    651 , &quot;Practical exploitation on system vulnerability of ProtoGENI.&quot;
    652 Proceedings of the 49th Annual Southeast Regional Conference, Kennesaw, Georgia, ACM, New York, NY, USA,
    653 2011.
    654 doi:10.1145/2016039.2016073.
    655 <a href="http://dx.doi.org/10.1145/2016039.2016073">http://dx.doi.org/10.1145/2016039.2016073</a>
    656 <br><br><b>Abstract: </b>Global Environment for Network Innovations (GENI) is a unique virtual laboratory for at-scale networking experimentation exploring future Internets. The successful development of GENI has to consider security problems from the design and prototyping stages. However, in many cases, system vulnerability cannot be found unless through real experimentation bearing purposeful and meaningful designs. In this paper, we introduce some of our efforts in exploring the security vulnerabilities in ProtoGENI, a prototype implementation and deployment of GENI. Our results show potential breach on security of GENI in terms of availability. We make suggestions on potential defense strategies in order to improve the ProtoGENI security and its development.
    657 </li>
    658 <br>
    659 
    660 
    661 
    662 <li>
    663 <b>Li, Dawei and Hong, Xiaoyan and Bowman, Jason</b>
    664 , &quot;Evaluation of Security Vulnerabilities by Using ProtoGENI as a Launchpad.&quot;
    665 IEEE Global Communications Conference (GLOBECOM 2011),
    666 2011.
    667 
    668 <a href="ftp://202.38.75.7/pub/&#x0025;D0&#x0025;C2&#x0025;CE&#x0025;C4&#x0025;BC&#x0025;FE&#x0025;BC&#x0025;D0&#x0025;20(2)/DATA/PID1102190.PDF">ftp://202.38.75.7/pub/&#x0025;D0&#x0025;C2&#x0025;CE&#x0025;C4&#x0025;BC&#x0025;FE&#x0025;BC&#x0025;D0&#x0025;20(2)/DATA/PID1102190.PDF</a>
    669 <br><br><b>Abstract: </b>In this paper we analyze the security architecture of ProtoGENI. ProtoGENI is a prototype control framework implementation of GENI (Global Environment for Network Innovations). We perform a variety of experiments in an effort to identify potential vulnerabilities presented in the current implementation. We classify our attacks into three types: data plane to data plane, data plane to control plane, and data plane to Internet. Our results indicate the potential for a breach of confidentiality and availability internally within ProtoGENI, as well as risks to external Internet. We make suggestions outlining possible defense strategies to improve ProtoGENI security and aid in future development
    670 </li>
    671 <br>
    672 
    673 
    674 
    675 <li>
    676 <b>Mandal, A. and Xin, Yufeng and Baldine, I. and Ruth, P. and Heerman, C. and Chase, J. and Orlikowski, V. and Yumerefendi, A.</b>
    677 , &quot;Provisioning and Evaluating Multi-domain Networked Clouds for Hadoop-based Applications.&quot;
    678 Cloud Computing Technology and Science (CloudCom), 2011 IEEE Third International Conference on,
    679 2011.
    680 doi:10.1109/CloudCom.2011.107.
    681 <a href="http://dx.doi.org/10.1109/CloudCom.2011.107">http://dx.doi.org/10.1109/CloudCom.2011.107</a>
    682 
    683 </li>
    684 <br>
    685 
    686 
    687 
    688 <li>
    689 <b>Ozcelik, Ilker and Brooks, Richard R.</b>
    690 , &quot;Security experimentation using operational systems.&quot;
    691 Proceedings of the Seventh Annual Workshop on Cyber Security and Information Intelligence Research, Oak Ridge, Tennessee, ACM, New York, NY, USA,
    692 2011.
    693 doi:10.1145/2179298.2179388.
    694 <a href="http://dx.doi.org/10.1145/2179298.2179388">http://dx.doi.org/10.1145/2179298.2179388</a>
    695 <br><br><b>Abstract: </b>Computers and Internet have evolved into necessary tools for our professional, personal and social lives. As a result of this growing dependence, there is a concern that these systems remain protected and available. This concern increases exponentially when considering systems such as smart power grids. Therefore, research should be conducted to develop effective ways of detecting system anomalies. To have realistic results, the studies should be tested on real systems. However, it is not possible to test these experiments on the live network. With the recent collaboration of Universities and research labs, a new experiment test bed has been established. As a result, experiments can now be implemented on real networks. In our study, we design an experiment to analyze Distributed Denial of Service Attacks (DDoS Attack) on a real network with real Internet traffic. The approach that we use in our study can easily be generalized to apply to smart power grids.
    696 </li>
    697 <br>
    698 
    699 
    700 
    701 <li>
    702 <b>Patali, Rohit</b>
    703 , &quot;Utility-Directed Resource Allocation in Virtual Desktop Clouds (Master's thesis).&quot;
    704 
    705 2011.
    706 
    707 <a href="https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1306872632">https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1306872632</a>
    708 <br><br><b>Abstract: </b>User communities are rapidly transitioning their ” traditional desktops” that have dedicated hardware and software installations into ” virtual desktop clouds” (VDCs) that are accessible via thin-clients. To allocate and manage VDC resources for Internet-scale desktop delivery, existing work focuses mainly on managing server-side resources based on utility functions of CPU and memory loads, and do not consider network health and thin-client user experience. Resource allocations without combined utility-directed information of system loads, network health and thin-client user experience in VDC platforms inevitably results in costly guesswork and overprovisioning of resources. In this thesis, an analytical model i.e., ” Utility-Directed Resource Allocation Model (U-RAM)” is presented to solve the combined utility-directed resource allocation problem within VDCs. The solution uses an iterative algorithm that leverages utility functions of system, network and human components obtained using a novel virtual desktop performance benchmarking toolkit i.e., ” VDBench”. The combined utility functions are used to direct decision schemes based on Kuhn-Tucker optimality conditions for creating user desktop pools and determining optimal resource allocation size/location. U-RAM is evaluated in a VDC testbed featuring: (a) popular user applications (Spreadsheet Calculator, Internet Browser, Media Player, Interactive Visualization), and (b) TCP/UDP based thin-client protocols (RDP, RGS, PCoIP) under a variety of user load and network health conditions. Evaluation results demonstrate that U-RAM solution maximizes VDC scalability i.e., 'VDs per core density', and 'user connections quantity', while delivering satisfactory thin-client user experience.
    709 </li>
    710 <br>
    711 
    712 
    713 
    714 <li>
    715 <b>Paul, Subharthi and Pan, Jianli and Jain, Raj</b>
    716 , &quot;Architectures for the future networks and the next generation Internet: A survey.&quot;
    717 Computer Communications, Elsevier Science Publishers B. V., Amsterdam, The Netherlands, The Netherlands,
    718 2011.
    719 doi:10.1016/j.comcom.2010.08.001.
    720 <a href="http://dx.doi.org/10.1016/j.comcom.2010.08.001">http://dx.doi.org/10.1016/j.comcom.2010.08.001</a>
    721 <br><br><b>Abstract: </b>Networking research funding agencies in USA, Europe, Japan, and other countries are encouraging research on revolutionary networking architectures that may or may not be bound by the restrictions of the current TCP/IP based Internet. We present a comprehensive survey of such research projects and activities. The topics covered include various testbeds for experimentations for new architectures, new security mechanisms, content delivery mechanisms, management and control frameworks, service architectures, and routing mechanisms. Delay/disruption tolerant networks which allow communications even when complete end-to-end path is not available are also discussed.
    722 </li>
    723 <br>
    724 
    725 
    726 
    727 <li>
    728 <b>Quan, John and Nance, Kara and Hay, Brian</b>
    729 , &quot;A Mutualistic Security Service Model: Supporting Large-Scale Virtualized Environments.&quot;
    730 IT Professional,
    731 2011.
    732 doi:10.1109/MITP.2011.36.
    733 <a href="http://dx.doi.org/10.1109/MITP.2011.36">http://dx.doi.org/10.1109/MITP.2011.36</a>
    734 <br><br><b>Abstract: </b>Applying a mutualistic security service model to large-scale virtualized environments that rely on contributed hardware lets researchers improve security in exchange for resources. The authors discuss this model in the context of the Global Environment for Network Innovation (GENI) project.
    735 </li>
    736 <br>
    737 
    738 
    739 
    740 <li>
    741 <b>Rohrer, Justin P. and &#x43;&#x0327;etinkaya, Egemen K. and Sterbenz, James P. G.</b>
    742 , &quot;Progress and challenges in large-scale future internet experimentation using the GpENI programmable testbed.&quot;
    743 Proceedings of the 6th International Conference on Future Internet Technologies, Seoul, Republic of Korea, ACM, New York, NY, USA,
    744 2011.
    745 doi:10.1145/2002396.2002409.
    746 <a href="http://dx.doi.org/10.1145/2002396.2002409">http://dx.doi.org/10.1145/2002396.2002409</a>
    747 <br><br><b>Abstract: </b>GpENI is evolving to provide a promising environment in which to do experimental research in the resilience and survivability of future networks, by allowing programmable control over topology and mechanism, while providing the scale and global reach needed to conduct network experiments far beyond the capabilities of a conventional testbed. Addressing this need at scale introduces a number of challenges both in deployment and in collecting results that can be directly compared to simulation results for cross-verification purposes. In this short paper we present the scope, design goals, challenges, and current status of the GpENI programmable testbed, as well as an overview and examples of the types of experiments we are beginning to run.
    748 </li>
    749 <br>
    750 
    751 
    752 
    753 <li>
    754 <b>Seskar, Ivan and Nagaraja, Kiran and Nelson, Sam and Raychaudhuri, Dipankar</b>
    755 , &quot;MobilityFirst future internet architecture project.&quot;
    756 Proceedings of the 7th Asian Internet Engineering Conference, Bangkok, Thailand, ACM, New York, NY, USA,
    757 2011.
    758 doi:10.1145/2089016.2089017.
    759 <a href="http://dx.doi.org/10.1145/2089016.2089017">http://dx.doi.org/10.1145/2089016.2089017</a>
    760 <br><br><b>Abstract: </b>This short paper presents an overview of the MobilityFirst network architecture, which is a clean-slate project being conducted as part of the NSF Future Internet Architecture (FIA) program. The proposed architecture is intended to directly address the challenges of wireless access and mobility at scale, while also providing new multicast, anycast, multi-path and context-aware services needed for emerging mobile Internet application scenarios. Key protocol components of the proposed architecture are: (a) separation of naming from addressing; (b) public key based self-certifying names (called globally unique identifiers or GUIDs) for network-attached objects; (c) global name resolution service (GNRS) for dynamic name-to-address binding; (d) delay-tolerant and storage-aware routing (GSTAR) capable of dealing with wireless link quality fluctuations and disconnections; (e) hop-by-hop transport of large protocol data units; and (f) location or context-aware services. The basic operations of a MobilityFirst router are outlined. This is followed by a discussion of ongoing proof-of-concept prototyping and experimental evaluation efforts for the MobilityFirst protocol stack. In conclusion, a brief description of an ongoing multi-site experimental deployment of the MobilityFirst protocol stack on the GENI testbed is provided.
    761 </li>
    762 <br>
    763 
    764 
    765 
    766 <li>
    767 <b>Shen, Haiying and Liu, Guoxin</b>
    768 , &quot;Harmony: Integrated Resource and Reputation Management for Large-Scale Distributed Systems.&quot;
    769 2011 Proceedings of 20th International Conference on Computer Communications and Networks (ICCCN), Lahaina, HI, USA, IEEE,
    770 2011.
    771 doi:10.1109/ICCCN.2011.6005739.
    772 <a href="http://dx.doi.org/10.1109/ICCCN.2011.6005739">http://dx.doi.org/10.1109/ICCCN.2011.6005739</a>
    773 <br><br><b>Abstract: </b>Advancements in technology over the past decade are leading to a promising future for large-scale distributed systems, where globally-scattered distributed resources are collectively pooled and used in a cooperative manner to achieve unprecedented petascale supercomputing capabilities. The issues of resource management (resMgt) and reputation management (repMgt) need to be addressed in order to ensure the successful deployment of large-scale distributed systems. However, these two issues have typically been addressed separately, despite the significant interdependencies between them: resMgt needs repMgt to provide a cooperative environment for resource sharing, and in turn facilitates repMgt to evaluate multi-faceted node reputations for providing different resources. Current repMgt methods provide a single reputation value for each node in providing all types of resources. However, a node willing to provide one resource may not be willing to provide another resource. In addition, current repMgt methods often guide node selection policy to select the highest-reputed nodes, which may overload these nodes. Also, few works exploited node reputation in resource selection in order to fully and fairly utilize resources in the system and to meet users' diverse QoS demands. We propose a system called Harmony that integrates resMgt and repMgt in a harmonious manner. Harmony incorporates two key innovations: integrated multi-faceted resource/reputation management and multi-QoS-oriented resource selection. The trace data we collected from an online trading platform confirms the importance of multi-faceted reputation and potential problems with highest-reputed node selection. Trace-driven experiments performed on PlanetLab show that Harmony outperforms existing resMgt and repMgt in terms of the success rate, service delay, and efficiency.
    774 </li>
    775 <br>
    776 
    777 
    778 
    779 <li>
    780 <b>Sridharan, Mukundan and Calyam, Prasad and Venkataraman, Aishwarya and Berryman, Alex</b>
    781 , &quot;Defragmentation of Resources in Virtual Desktop Clouds for Cost-Aware Utility-Optimal Allocation.&quot;
    782 2011 Fourth IEEE International Conference on Utility and Cloud Computing, Melbourne, Australia, IEEE,
    783 2011.
    784 doi:10.1109/UCC.2011.41.
    785 <a href="http://dx.doi.org/10.1109/UCC.2011.41">http://dx.doi.org/10.1109/UCC.2011.41</a>
    786 <br><br><b>Abstract: </b>Cloud Service Providers (CSPs) make virtual desktop cloud (VDC) resource provisioning decisions within desktop pools based on user groups and their application profiles. Such provisioning is aimed to satisfy acceptable user quality of experience (QoE) levels and is coupled with subsequent placement of VDs across distributed data centers. The placement decisions are influenced by session latency, load balancing and operation cost constraints. In this paper, we identify the resource fragmentation problem that occurs when placement is done opportunistically to minimize provisioning time and deliver satisfactory user QoE. To solve this problem, which inherently is an NP-Hard problem, we propose a defragmentation scheme that has fast convergence time and has three levels of complexity: (i) &#x75;&#x0308;tility fair provisioning&#x20;&#x0308;(UFP) to optimize resource provisioning within a data center - to achieve relative fairness between desktop pools, (ii) &#x73;&#x0308;tatic migration-free utility optimal placement and provisioning&#x20;&#x0308;(MUPP) to optimize resource provisioning between multiple data centers - to improve performance, and (iii) &#x64;&#x0308;ynamic global utility optimal placement and provisioning&#x20;&#x0308;(GUPP) to optimize resource provisioning using cost-aware and utility-maximal VD re-allocations and migrations - to increase scalability. We evaluate our defragmentation scheme against 'least latency', 'least load', and 'least cost' schemes using a novel &#x56;&#x0308;DC-Sim&#x20;&#x0308;simulator that we have developed in this study. Our simulations leverage profiles of user groups and their applications within desktop pools, obtained from a real VDC test bed. Our simulation results demonstrate that defragmentation is an important optimization step that can enable CSPs to achieve fairness, substantially improve user QoE and increase VDC scalability.
    787 </li>
    788 <br>
    789 
    790 
    791 
    792 <li>
    793 <b>Sterbenz, J. P. G. and Egemen and Hameed, M. A. and Jabbar, A. and Rohrer, J. P.</b>
    794 , &quot;Modelling and analysis of network resilience.&quot;
    795 2011 Third International Conference on Communication Systems and Networks (COMSNETS 2011), Bangalore, IEEE,
    796 2011.
    797 doi:10.1109/COMSNETS.2011.5716502.
    798 <a href="http://dx.doi.org/10.1109/COMSNETS.2011.5716502">http://dx.doi.org/10.1109/COMSNETS.2011.5716502</a>
    799 <br><br><b>Abstract: </b>As the Internet becomes increasingly important to all aspects of society, the consequences of disruption become increasingly severe. Thus it is critical to increase the resilience and survivability of the future network. We define resilience as the ability of the network to provide desired service even when challenged by attacks, large-scale disasters, and other failures. This paper describes a comprehensive methodology to evaluate network resilience using a combination of analytical and simulation techniques with the goal of improving the resilience and survivability of the Future Internet.
    800 </li>
    801 <br>
    802 
    803 
    804 
    805 <li>
    806 <b>Tiako, Pierre F.</b>
    807 , &quot;Perspectives of delegation in team-based distributed software development over the GENI infrastructure (NIER track).&quot;
    808 Proceedings of the 33rd International Conference on Software Engineering, Waikiki, Honolulu, HI, USA, ACM, New York, NY, USA,
    809 2011.
    810 doi:10.1145/1985793.1985905.
    811 <a href="http://dx.doi.org/10.1145/1985793.1985905">http://dx.doi.org/10.1145/1985793.1985905</a>
    812 <br><br><b>Abstract: </b>Team-based distributed software development (TBDSD) is one of the single biggest challenges facing software companies. The need to manage development efforts and resources in different locations increase the complexity and cost of modern day software development. Current software development environments do not provide suitable support to delegate task among teams with appropriate directives. TBDSD is also limited to the current internet capabilities. One of the resulting problems is the difficulty to delegate and control tasks assigned among remote teams. This paper proposes (1) a new framework for delegation in TBDSD, and (2) perspectives for deploying Process-centered Software Engineering Environments (PSEE) over the Global Environment for Network Innovations (GENI) infrastructure. GENI, the 'future Internet' that is taking shape in prototypes across the US, will allow, in the context of our study, to securely access and share software artifacts, resources, and tools as never before seen over the current Internet.
     3049<b>Valancius, Vytautas and Ravi, Bharath and Feamster, Nick and Snoeren, Alex C.</b>
     3050, &quot;Quantifying the benefits of joint content and network routing.&quot;
     3051Proceedings of the ACM SIGMETRICS/international conference on Measurement and modeling of computer systems - SIGMETRICS '13, Pittsburgh, PA, USA, ACM Press,
     30522013.
     3053doi:10.1145/2465529.2465762.
     3054<a href="http://dx.doi.org/10.1145/2465529.2465762">http://dx.doi.org/10.1145/2465529.2465762</a>
     3055<br><br><b>Abstract: </b>Online service providers aim to provide good performance for an increasingly diverse set of applications and services. One of the most effective ways to improve service performance is to replicate the service closer to the end users. Replication alone, however, has its limits: while operators can replicate static content, wide-scale replication of dynamic content is not always feasible or cost effective. To improve the latency of such services many operators turn to Internet traffic engineering. In this paper, we study the benefits of performing replica-to-end-user mappings in conjunction with active Internet traffic engineering. We present the design of PECAN, a system that controls both the selection of replicas (&#x63;&#x0308;ontent routing&#x29;&#x0308; and the routes between the clients and their associated replicas (&#x6e;&#x0308;etwork routing&#x29;&#x0308;. We emulate a replicated service that can perform both content and network routing by deploying PECAN on a distributed testbed. In our testbed, we see that jointly performing content and network routing can reduce round-trip latency by 4.3&#x0025; on average over performing content routing alone (potentially reducing service response times by tens of milliseconds or more) and that most of these gains can be realized with no more than five alternate routes at each replica.
    8133056</li>
    8143057<br>
     
    8273070<br>
    8283071
     3072<li>
     3073<b>Van Vorst, N. and Erazo, M. and Liu, J.</b>
     3074, &quot;PrimoGENI for hybrid network simulation and emulation experiments in GENI.&quot;
     3075Journal of Simulation,
     30762012.
     3077doi:10.1057/jos.2012.5.
     3078<a href="http://dx.doi.org/10.1057/jos.2012.5">http://dx.doi.org/10.1057/jos.2012.5</a>
     3079<br><br><b>Abstract: </b>The Global Environment for Network Innovations (GENI) is a community-driven research and development effort to build a collaborative and exploratory network experimentation platform—a 'virtual laboratory' for the design, implementation, and evaluation of future networks. The PrimoGENI project enables real-time network simulation by extending an existing network simulator to become part of the GENI federation to support large-scale experiments involving physical, simulated, and emulated network entities. In this paper, we describe a novel design of PrimoGENI, which aims at supporting realistic, scalable, and flexible network experiments with real-time simulation and emulation capabilities. We present a flexible emulation infrastructure that allows both remote client machines, local cluster nodes running virtual machines, and external networks to seamlessly interoperate with the simulated network running within a designated 'slice' of resources. We present the results of our preliminary validation and performance studies to demonstrate the capabilities as well as limitations of our approach.
     3080</li>
     3081<br>
     3082
    8293083
    8303084
     
    8433097
    8443098<li>
     3099<b>Van Vorst, N. and Liu, J.</b>
     3100, &quot;Realizing Large-Scale Interactive Network Simulation via Model Splitting.&quot;
     3101Principles of Advanced and Distributed Simulation (PADS), 2012 ACM/IEEE/SCS 26th Workshop on, IEEE,
     31022012.
     3103doi:10.1109/pads.2012.35.
     3104<a href="http://dx.doi.org/10.1109/pads.2012.35">http://dx.doi.org/10.1109/pads.2012.35</a>
     3105<br><br><b>Abstract: </b>This paper presents the model splitting method for large-scale interactive network simulation, which addresses the separation of concerns between network researchers, who focus on developing complex network models and conducting large-scale network experiments, and simulator developers, who are concerned with developing efficient simulation engines to achieve the best performance on parallel platforms. Modeling splitting divides the system into an interactive model to support user interaction, and an execution model to facilitate parallel processing. We describe techniques to maintain consistency and real-time synchronization between the two models. We also provide solutions to reduce the memory complexity of large network models and to ensure data persistency and access efficiency for out-of-core processing.
     3106</li>
     3107<br>
     3108
     3109
     3110
     3111<li>
     3112<b>Velusamy, G. and Gurkan, D. and Narayan, S. and Baily, S.</b>
     3113, &quot;Fault-Tolerant OpenFlow-Based Software Switch Architecture with LINC Switches for a Reliable Network Data Exchange.&quot;
     3114Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     31152014.
     3116doi:10.1109/gree.2014.17.
     3117<a href="http://dx.doi.org/10.1109/gree.2014.17">http://dx.doi.org/10.1109/gree.2014.17</a>
     3118<br><br><b>Abstract: </b>The switches are essential for forwarding the packets in a local area network. If a switch fails, then the packets are not able to reach their destination, in spite of their long journey from the source. The new trend in Software Defined Networking (SDN) has made the use of software switches such as the OpenvSwitch quite popular. These software switches are used in data centers to connect virtual machines on which application servers are deployed. Such switches have the advantages of software: ease of development and flexibility, with less optimal testing and reliability measures than hardware systems. The Software switches are required to be resilient to failure because the applications servers which are running from the VMs which are connected through them should always be connected with its clients. So fault-tolerance becomes an important aspect in the use of software switches. In this paper, we explore one mechanism for fault tolerance of LINC (Link Is Not Closed), an open source OpenFlow switch, which is written in Erlang programming language. Distributed system, concurrency, and fault-tolerance are built-in features of Erlang. We leverage these features of Erlang to realize a fault-tolerant distributed LINC switch system.
     3119</li>
     3120<br>
     3121
     3122
     3123
     3124<li>
     3125<b>Velusamy, Gandhimathi</b>
     3126, &quot;OpenFlow-based Distributed and Fault-Tolerant Software Switch Architecture (Master's thesis).&quot;
     3127
     31282014.
     3129
     3130<a href="http://repositories.tdl.org/uh-ir/bitstream/handle/10657/693/VELUSAMY-THESIS-2014.pdf">http://repositories.tdl.org/uh-ir/bitstream/handle/10657/693/VELUSAMY-THESIS-2014.pdf</a>
     3131<br><br><b>Abstract: </b>We are living in the era where each of us is connected with each other virtually across the globe. We are sharing the information electronically over the internet every second of our day. There are many networking devices involved in sending the information over the internet. They are routers, gateways, switches, PCs, laptops, handheld devices, etc. The switches are very crucial elements in delivering packets to the intended recipients. Now the networking field is moving towards Software Defined Networking and the network elements are being slowly replaced by the software applications run by OpenFlow protocols. For example the switching functionality in local area networks could be achieved with software switches like OpenvSwitch (OVS), LINC-Switch, etc. Now a days the organizations depend on the datacenters to run their services. The application servers are being run from virtual machines on the hosts to better utilize the computing resources and make the system more scalable. The application servers need to be continuously available to run the business for which they are deployed for. Software switches are used to connect virtual machines as an alternative to Top of Rack switches. If such software switch fails then the application servers will not be able to connect to its clients. This may severely impact the business serviced by the application servers, deployed on the virtual machines. For reliable data connectivity, the switching elements need to be continuously functional. There is a need for reliable and robust switches to cater the today's networking infrastructure. In this study, the software switch LINC-Switch is implemented as distributed application on multiple nodes to make it resilient to failure. The fault-tolerance is achieved by using the distribution properties of the programming language Erlang. By implementing the switch on three redundant nodes and starting the application as a distributed application, the switch will be serving its purpose very promptly by restarting it on other node in case it fails on the current node by using failover/takeover mechanisms of Erlang. The tolerance to failure of the LINC-Switch is verified with Ping based experiment on the GENI test bed and on the Xen-cluster in our Lab.
     3132</li>
     3133<br>
     3134
     3135
     3136
     3137<li>
     3138<b>Venkataraman, Aishwarya</b>
     3139, &quot;Defragmentation of Resources in Virtual Desktop clouds for Cost-aware Utility-maximal Allocation (Master's thesis).&quot;
     3140
     31412012.
     3142
     3143<a href="https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1339747492">https://etd.ohiolink.edu/!etd.send&#x005F;file?accession=osu1339747492</a>
     3144<br><br><b>Abstract: </b>Cloud Service Providers (CSPs) make virtual desktop cloud (VDC) resource provisioning decisions within desktop pools based on user groups and their application pro- files. Such provisioning is aimed to not only satisfy acceptable user quality of experience (QoE) levels and provide high scalability, but also provide ” knobs” to CSPs to operate according their economic policies. The next challenge is to place user VD requests in an optimal and fast manner across distributed data centers. The placement decisions are influenced by session latency, load balancing and operation cost constraints. In this work, we identify the resource fragmentation problem that occurs when placement is done opportunistically to minimize provisioning time and deliver satisfactory user QoE. To solve this problem, which inherently is an NP-Hard problem, we propose a defragmentation scheme that has fast convergence time and has three levels of complexity: (i) ” Economics-directed resource allocation model” (E-RAM) that considers economic policies while optimizing resource provisioning within a data center (ii) ” Cost-aware Utility-maximal Local Placement” to optimize resource provisioning between multiple data centers, and (iii) ” Costaware Utility-maximal Global Placement with Migration” to optimize resource provisioning using cost-aware and utility-maximal VD re-allocations and migrations - to increase scalability and performance. We evaluate our E-RAM, Cost-aware Utility-maximal Local and Global Placement schemes using a novel ” VDC-Sim” simulator that we have developed in this study. Our simulations leverage profiles of user groups and their applications within desktop pools, obtained from a real VDC testbed. We also implemented our schemes in a real cloud infrastructure. Our results demonstrate that defragmentation is an important optimization step and defragmentation together with E-RAM and our Cost-aware Utilitymaximal placement schemes can enable CSPs to achieve optimal user QoE, higher VDC scalability, improved system performance and resilience.
     3145</li>
     3146<br>
     3147
     3148
     3149
     3150<li>
     3151<b>Vulimiri, Ashish and Michel, Oliver and Godfrey, P. Brighten and Shenker, Scott</b>
     3152, &quot;More is Less: Reducing Latency via Redundancy.&quot;
     3153Proceedings of the 11th ACM Workshop on Hot Topics in Networks, Redmond, Washington, ACM, New York, NY, USA,
     31542012.
     3155doi:10.1145/2390231.2390234.
     3156<a href="http://dx.doi.org/10.1145/2390231.2390234">http://dx.doi.org/10.1145/2390231.2390234</a>
     3157<br><br><b>Abstract: </b>Low latency is critical for interactive networked applications. But while we know how to scale systems to increase capacity, reducing latency --- especially the tail of the latency distribution --- can be much more difficult. We argue that the use of redundancy in the context of the wide-area Internet is an effective way to convert a small amount of extra capacity into reduced latency. By initiating redundant operations across diverse resources and using the first result which completes, redundancy improves a system's latency even under exceptional conditions. We demonstrate that redundancy can significantly reduce latency for small but critical tasks, and argue that it is an effective general-purpose strategy even on devices like cell phones where bandwidth is relatively constrained.
     3158</li>
     3159<br>
     3160
     3161
     3162
     3163<li>
    8453164<b>Wallace, Scott A. and Muhammad, Monzur and Mache, Jens and Cappos, Justin</b>
    8463165, &quot;Hands-on Internet with Seattle and Computers from Across the Globe.&quot;
     
    8503169<a href="http://portal.acm.org/citation.cfm?id=2037151.2037181">http://portal.acm.org/citation.cfm?id=2037151.2037181</a>
    8513170<br><br><b>Abstract: </b>The Internet Connectivity module is a short assignment covering distributed computing and networking. The Internet Connectivity module is part of the curriculum created for the Northwest Distributed Computer Science Department and is built upon the Seattle distributed computing platform. In this paper, we describe the module and illustrate how Seattle facilitates networking projects and experiments that use computers/resources from across the globe. In addition, we describe how the Internet Connectivity module was used in two courses, provide some comments on students' reactions to the project, and conclude with suggestions for faculty considering how to use this module in their future courses.
     3171</li>
     3172<br>
     3173
     3174
     3175
     3176<li>
     3177<b>Wang, Han and Lee, Ki S. and Li, Erluo and Lim, Chiun L. and Tang, Ao and Weatherspoon, Hakim</b>
     3178, &quot;Timing is Everything: Accurate, Minimum Overhead, Available Bandwidth Estimation in High-speed Wired Networks.&quot;
     3179Proceedings of the 2014 Conference on Internet Measurement Conference, Vancouver, BC, Canada, ACM, New York, NY, USA,
     31802014.
     3181doi:10.1145/2663716.2663746.
     3182<a href="http://dx.doi.org/10.1145/2663716.2663746">http://dx.doi.org/10.1145/2663716.2663746</a>
     3183<br><br><b>Abstract: </b>Active end-to-end available bandwidth estimation is intrusive, expensive, inaccurate, and does not work well with bursty cross traffic or on high capacity links. Yet, it is important for designing high performant networked systems, improving network protocols, building distributed systems, and improving application performance. In this paper, we present minProbe which addresses unsolved issues that have plagued available bandwidth estimation. As a middlebox, minProbe measures and estimates available bandwidth with high-fidelity, minimal-cost, and in userspace; thus, enabling cheaper (virtually no overhead) and more accurate available bandwidth estimation. MinProbe performs accurately on high capacity networks up to 10 Gbps and with bursty cross traffic. We evaluated the performance and accuracy of minProbe over a wide-area network, the National Lambda Rail (NLR), and within our own network testbed. Results indicate that minProbe can estimate available bandwidth with error typically no more than 0.4 Gbps in a 10 Gbps network.
     3184</li>
     3185<br>
     3186
     3187
     3188
     3189<li>
     3190<b>Wang, K. C. and Brinn, M. and Mambretti, J.</b>
     3191, &quot;From federated software defined infrastructure to future internet architecture.&quot;
     3192Science and Technology Conference (Modern Networking Technologies) (MoNeTeC), 2014 First International, IEEE,
     31932014.
     3194doi:10.1109/monetec.2014.6995605.
     3195<a href="http://dx.doi.org/10.1109/monetec.2014.6995605">http://dx.doi.org/10.1109/monetec.2014.6995605</a>
     3196<br><br><b>Abstract: </b>Significant efforts have been devoted to creating large scale compute and network testbeds for studying future Internet challenges. Besides large geographic span, the common emphasis is programmability, allowing researchers to reserve or create, via software, flexible sets of compute and network resources over specified topologies to execute research prototypes of new protocols, processes, and applications. Also emphasized are virtualization, instrumentation, and software defined networking (SDN) capabilities of the infrastructure. SDN in particular stimulated significant interests in academia, industry, and public sectors to re-imagine the future computing and networking infrastructure landscape and roadmap while it becomes increasingly utilized in production environments. Amidst these interests, one can start to capture desirable characteristics to glimpse the potential architecture of the future Internet. In this paper, we discuss the significance of compute-network interaction across complex, highly customized federated architecture in the future Internet. Infrastructure federation has been happening across multiple dimensions. Federation expands the scope of infrastructure, geographically and administratively, for use by members of different organizations. For example, federation initiatives are underway among: 1) US Global Environment for Network Innovations (GENI), Europe Future Internet Research and Experimentation (FIRE), and future Internet testbeds in Asia, South America, and Canada, 2) university production infrastructure, 3) US cities, 4) US public research institutes, and 5) commercial infrastructure. While requirements and objectives differ, they must all address a common set of issues. Such federation suggests the fundamental needs of applications to interact with compute and network resources across a generic, federated, future Internet environment.
     3197</li>
     3198<br>
     3199
     3200
     3201
     3202<li>
     3203<b>Wang, Qing and Xu, Ke and Izard, Ryan and Kribbs, Benton and Porter, Joseph and Wang, Kuang-Ching and Prakash, Aditya and Ramanathan, Parmesh</b>
     3204, &quot;GENI Cinema: An SDN-Assisted Scalable Live Video Streaming Service.&quot;
     3205Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     32062014.
     3207doi:10.1109/icnp.2014.84.
     3208<a href="http://dx.doi.org/10.1109/icnp.2014.84">http://dx.doi.org/10.1109/icnp.2014.84</a>
     3209<br><br><b>Abstract: </b>This paper introduces GENI Cinema (GC), a system that provides a scalable live video streaming service based on dynamic traffic steering with software defined networking (SDN) and demand driven instantiation of video relay servers in NSF GENI's distributed cloud environments. While the service can be used to relay a multitude of video content, its initial objective is to support live video streaming of educational content such as lectures and seminars among university campuses. Users on any campus would bootstrap video upload or download via a public web portal and, for scalability, have the video delivered seamlessly across the network over one or multiple paths selected and dynamically controlled by GC. The architecture aims to provide a framework for addressing several well-known limitations of video streaming in today's Internet, where little control is available for controlling forwarding paths of on demand live video streams. GC utilizes GENI's distributed cloud servers to host on-demand video servers/relays and its Open Flow SDN to achieve seamless video upload/download and optimization of forwarding paths in the network core. This paper presents the architecture and an early prototype of the basic GC framework, together with some initial performance measurement results.
     3210</li>
     3211<br>
     3212
     3213
     3214
     3215<li>
     3216<b>Wang, Yuefeng and Akhtar, Nabeel and Matta, Ibrahim</b>
     3217, &quot;Programming Routing Policies for Video Traffic.&quot;
     3218Network Protocols (ICNP), 2014 IEEE 22nd International Conference on, IEEE,
     32192014.
     3220doi:10.1109/icnp.2014.80.
     3221<a href="http://dx.doi.org/10.1109/icnp.2014.80">http://dx.doi.org/10.1109/icnp.2014.80</a>
     3222<br><br><b>Abstract: </b>Making the network programmable simplifies network management and enables network innovations. The Recursive Inter Network Architecture (RINA) is our solution to enable network programmability. ProtoRINA is a user-space prototype of RINA and provides users with a framework with common mechanisms so a user can program recursive-networking policies without implementing mechanisms from scratch. In this paper, we focus on how routing policies, which is an important aspect of network management, can be programmed using ProtoRINA, and demonstrate how ProtoRINA can be used to achieve better performance for a video streaming application by instantiating different routing policies over the GENI (Global Environment for Network Innovations) test bed, which provides a large-scale experimental facility for networking research.
     3223</li>
     3224<br>
     3225
     3226
     3227
     3228<li>
     3229<b>Wang, Yuefeng and Esposito, F. and Matta, I.</b>
     3230, &quot;Demonstrating RINA Using the GENI Testbed.&quot;
     3231Research and Educational Experiment Workshop (GREE), 2013 Second GENI, IEEE,
     32322013.
     3233doi:10.1109/gree.2013.26.
     3234<a href="http://dx.doi.org/10.1109/gree.2013.26">http://dx.doi.org/10.1109/gree.2013.26</a>
     3235<br><br><b>Abstract: </b>The inability of the current Internet architecture to accommodate modern requirements has spurred novel designs for future Internet architectures. The Global Environment for Network Innovations (GENI) is a wide-area virtual network testbed which allows experimentation of such architectures for possible deployment. We have contributed to the efforts of redesigning the Internet with a Recursive InterNetwork Architecture (RINA), and in this paper we demonstrate its practicability by running a prototype on the GENI testbed. We focus on testing two fundamental features of our architecture: security and manageability, discussing in detail how the experimentation was carried, and pointing out some lessons learned using the testbed.
     3236</li>
     3237<br>
     3238
     3239
     3240
     3241<li>
     3242<b>Wang, Yuefeng and Matta, I. and Akhtar, N.</b>
     3243, &quot;Experimenting with Routing Policies Using ProtoRINA over GENI.&quot;
     3244Research and Educational Experiment Workshop (GREE), 2014 Third GENI, IEEE,
     32452014.
     3246doi:10.1109/gree.2014.11.
     3247<a href="http://dx.doi.org/10.1109/gree.2014.11">http://dx.doi.org/10.1109/gree.2014.11</a>
     3248<br><br><b>Abstract: </b>ProtoRINA is a user-space prototype of the Recursive InterNetwork Architecture (RINA), a new architecture that overcomes inherent weaknesses of the current Internet, e:g:, security, mobility, and manageability. By separating mechanisms and policies, RINA supports the programmability of different control and management policies over different communication scopes while using the same mechanisms. GENI (Global Environment for Network Innovations) provides a large-scale virtual network testbed that supports experimentation and possible deployment of future network architectures. In this paper, using ProtoRINA over GENI resources, we demonstrate how RINA's support for the scoping of routing control and management, and instantiation of different routing policies, can be leveraged to yield faster convergence and lower routing overhead in the face of node or link failures.
     3249</li>
     3250<br>
     3251
     3252
     3253
     3254<li>
     3255<b>Willner, Alexander and Magedanz, Thomas</b>
     3256, &quot;FIRMA: A Future Internet resource management architecture.&quot;
     3257Teletraffic Congress (ITC), 2014 26th International, IEEE,
     32582014.
     3259doi:10.1109/itc.2014.6932981.
     3260<a href="http://dx.doi.org/10.1109/itc.2014.6932981">http://dx.doi.org/10.1109/itc.2014.6932981</a>
     3261<br><br><b>Abstract: </b>The Internet is broken and there are several approaches to fix it. In order to validate the different attempts, they need to be evaluated within large-scale environments involving numerous heterogeneous resources. As a result, several testbeds have been established along with a number of competitive mechanisms to federate them. Since most of these protocols try to address similar issues, combining and unifying them is subject of current research. This leads to a complex environment for testbed owners and developers. Furthermore, it is foreseeable that even more federation approaches in different application domains will emerge in the future. Therefore, we propose an extensible architecture that allows to be federation protocol agnostic. The fundamental idea is to allow interoperability on the level of a semantic information model and to separate delivery mechanism specific implementations from a common core. The requirements for such an architecture have been extracted from latest European Future Internet research projects and its practicability is being evaluated by an initial implementation.
     3262</li>
     3263<br>
     3264
     3265
     3266
     3267<li>
     3268<b>Wong, G. and Ricci, R. and Duerig, J. and Stoller, L. and Chikkulapelly, S. and Seok, Woojin</b>
     3269, &quot;Partitioning Trust in Network Testbeds.&quot;
     3270System Science (HICSS), 2012 45th Hawaii International Conference on, IEEE,
     32712012.
     3272doi:10.1109/HICSS.2012.466.
     3273<a href="http://dx.doi.org/10.1109/HICSS.2012.466">http://dx.doi.org/10.1109/HICSS.2012.466</a>
     3274<br><br><b>Abstract: </b>Traditionally, test beds for networking and systems research have been designed as monolithic facilities: they contain a single root of trust. The resources in the facility are assumed to be administered by a single entity or a set of mutually-trusting entities. All user management, including vouching for users' identities and taking responsibility for their actions, is done using a flat trust structure or a simple hierarchy with the facility itself as the root. This design is not a good match for test beds that are composed of multiple autonomous facilities, or in which different parts of the test bed operate under different trust models. In this paper, we argue that partitioned trust is increasingly important in large scale and security-sensitive test beds. We present a design that accomplishes this partitioning by using multiple trust roots. The trust domains created by these roots may decide, independently, how much trust to place in each other, and can apply policies based on the domain or principal that originates a request. The domains could represent separately administered facilities (as in a federated test bed), or they could represent sections within a single facility that run with different trust models (for example, with differing levels of security.) We have implemented this design in ProtoGENI, a control framework for federated test beds, we include details of this implementation and share experiences from using it in an active deployment with hundreds of users.
     3275</li>
     3276<br>
     3277
     3278
     3279
     3280<li>
     3281<b>Xiao, Zhifeng and Fu, Bo and Xiao, Yang and Chen, C. L. Philip and Liang, Wei</b>
     3282, &quot;A review of GENI authentication and access control mechanisms.&quot;
     3283International Journal of Security and Networks,
     32842013.
     3285doi:10.1504/ijsn.2013.055046.
     3286<a href="http://dx.doi.org/10.1504/ijsn.2013.055046">http://dx.doi.org/10.1504/ijsn.2013.055046</a>
     3287<br><br><b>Abstract: </b>The purpose of this paper is to investigate the authentication and access control mechanisms for Global Environment Network Innovation (GENI). First, we will deliver an extensive survey of the existing authentication and access control techniques in general. We will then study how authentication and access control policies of GENI projects are implemented and how these mechanisms are integrated into the project control frameworks. Finally, we will summarise the advantages and disadvantages of the authentication and access control methods employed in GENI. We believe that the given review is valuable to those who are interested in the internal design of the current GENI security mechanisms.
     3288</li>
     3289<br>
     3290
     3291
     3292
     3293<li>
     3294<b>Xin, Yufeng and Baldin, Ilya and Chase, Jeff and Ogan, Kemafor</b>
     3295, &quot;Leveraging Semantic Web Technologies for Managing Resources in a Multi-Domain Infrastructure-as-a-Service Environment.&quot;
     3296CoRR,
     32972014.
     3298
     3299<a href="http://arxiv.org/abs/1403.0949">http://arxiv.org/abs/1403.0949</a>
     3300
     3301</li>
     3302<br>
     3303
     3304
     3305
     3306<li>
     3307<b>Xin, Yufeng and Baldin, Ilya and Heermann, Chris and Mandal, Anirban and Ruth, Paul</b>
     3308, &quot;Scaling up applications over distributed clouds with dynamic layer-2 exchange and broadcast service.&quot;
     3309Teletraffic Congress (ITC), 2014 26th International, IEEE,
     33102014.
     3311doi:10.1109/itc.2014.6932973.
     3312<a href="http://dx.doi.org/10.1109/itc.2014.6932973">http://dx.doi.org/10.1109/itc.2014.6932973</a>
     3313<br><br><b>Abstract: </b>In this paper, we study the problem of provisioning large-scale virtual clusters over federated clouds connected by multi-domain, layer-2 wide area networks. We first present the virtual cluster request abstraction and the abstraction models for substrate resource pools. Based on these two abstraction models, we developed a novel layer-2 exchange mechanism and an implementation of it in a multi-domain networked cloud environment. The design of the mechanism takes into consideration the realistic constraints in current network and cloud systems. We show that efficient cluster splitting, cloud data center selection and resource allocation algorithms can be developed to provision large-scale virtual clusters across cloud sites. A prototype system has been deployed and integrated into the ExoGENI testbed for about a year, and is being heavily used by scientific and data analytic applications.
     3314</li>
     3315<br>
     3316
     3317<li>
     3318<b>Xin, Yufeng and Baldin, Ilya and Heermann, Chris and Mandal, Anirban and Ruth, Paul</b>
     3319, &quot;Capacity of Inter-cloud Layer-2 Virtual Networking.&quot;
     3320Proceedings of the 2014 ACM SIGCOMM Workshop on Distributed Cloud Computing, Chicago, Illinois, USA, ACM, New York, NY, USA,
     33212014.
     3322doi:10.1145/2627566.2627573.
     3323<a href="http://dx.doi.org/10.1145/2627566.2627573">http://dx.doi.org/10.1145/2627566.2627573</a>
     3324<br><br><b>Abstract: </b>Due to the economy of scale of Ethernet networks and available dynamic circuit capability from the major national research and educational networks, VLAN (Virtual LAN) based virtual networking solution has been successfully adopted in some advanced distributed cloud systems. However, there are two major constraints in this adaptation: (1) dynamic circuit service is far from pervasive; (2) there is only limited VLAN tags offered by regional network service providers. In this paper, after examining layer-2 networking in large-scale distributed cloud environments, we present a graph theoretical model to study the network capacity in terms of the number of inter-cloud connections that can co-exist. We further design the algorithms to achieve this capacity for both point-to-point and multi-point inter-cloud connections in both static and dynamic scenarios. We also study a general topology embedding problem based on this model. As tagging is a common mechanism for isolating communication channels in other network layers, the proposed models and algorithms can be extended to optical and IP networks.
    8523325</li>
    8533326<br>
     
    8683341
    8693342
    870 <br>
    871 <a id="full-2012"><H2>GENI Publications for 2012</H2></a>
    872 
    873 
    874 <li>
    875 <b>Aikat, Jay and Hasan, Shaddi and Jeffay, Kevin and Smith, F. Donelson</b>
    876 , &quot;Discrete-Approximation of Measured Round Trip Time Distributions: A Model for Network Emulation.&quot;
    877 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    878 2012.
    879 
    880 
    881 <br><br><b>Abstract: </b>Empirical evaluations to study network performance, whether in a laboratory setting or on GENI testbeds, rely heavily on measurement-based modeling of round trip times (RTTs) to emulate realistic end-to-end delays of local and metropolitan area networks. For generating realistic traffic, we studied several models to emulate RTTs. In this paper, we performed experiments on real testbeds using synthetic TCP traffic generated from measurement data from a large university campus. As a result of our study, we present the Discrete- Approximation model for RTT (DA-RTT) emulation. Using three different metrics for performance evaluation, which include queue length at routers, connection response times, and connection durations, we demonstrate that the simple DA-RTT model closely represents the per-connection RTTs in the original traffic. While these experiments were performed in our laboratory, and not using GENI infrastructure, we present this as a possible model for adoption on GENI testbeds to emulate Round Trip Time Distributions for GENI experiments.
    882 </li>
    883 <br>
    884 
    885 
    886 
    887 <li>
    888 <b>Baldine, Ilia and Xin, Yufeng and Mandal, Anirban and Ruth, Paul and Yumerefendi, Aydan and Chase, Jeff</b>
    889 , &quot;ExoGENI: A Multi-Domain Infrastructure-as-a-Service Testbed.&quot;
    890 8th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TRIDENTCOM 2012),
    891 2012.
    892 
    893 
    894 <br><br><b>Abstract: </b>NSF's GENI program seeks to enable experiments that run within virtual network topologies built-to-order from testbed infrastructure offered by multiple providers (domains). GENI is often viewed as a network testbed integration effort, but behind it is an ambitious vision for multi-domain infrastructure-as-a-service (IaaS). This paper presents ExoGENI, a new GENI testbed that links GENI to two advances in virtual infrastructure services outside of GENI: open cloud computing (OpenStack) and dynamic circuit fabrics. ExoGENI orchestrates a federation of independent cloud sites and circuit providers through their native IaaS interfaces, and links them to other GENI tools and resources. The ExoGENI deployment consists of cloud site ``racks'' on host campuses within the US, linked with national research networks and other circuit networks through programmable exchange points. The ExoGENI sites and control software are enabled for software-defined networking using OpenFlow. ExoGENI offers a powerful unified hosting platform for deeply networked, multi-domain, multi-site cloud applications. We intend that ExoGENI will seed a larger, evolving platform linking other third-party cloud sites, transport networks, and other infrastructure services, and that it will enable real-world deployment of innovative distributed services and new visions of a Future Internet.
    895 </li>
    896 <br>
    897 
    898 
    899 
    900 <li>
    901 <b>Bavier, Andy and Coady, Yvonne and Mack, Tony and Matthews, Chris and Mambretti, Joe and McGeer, Rick and Mueller, Paul and Snoeren, Alex and Yuen, Marco</b>
    902 , &quot;GENICloud and transcloud.&quot;
    903 Proceedings of the 2012 workshop on Cloud services, federation, and the 8th open cirrus summit, San Jose, California, USA, ACM, New York, NY, USA,
    904 2012.
    905 doi:10.1145/2378975.2378980.
    906 <a href="http://dx.doi.org/10.1145/2378975.2378980">http://dx.doi.org/10.1145/2378975.2378980</a>
    907 <br><br><b>Abstract: </b>In this paper, we argue that federation of cloud systems requires a standard API for users to create, manage, and destroy virtual objects, and a standard naming scheme for virtual objects. We introduce an existing API for this purpose, the Slice-Based Federation Architecture, and demonstrate that it can be implemented on a number of existing cloud management systems. We introduce a simple naming scheme for virtual objects, and discuss its implementation.
    908 </li>
    909 <br>
    910 
    911 
    912 
    913 <li>
    914 <b>Bhanage, Gautam and Seskar, Ivan and Raychaudhuri, Dipankar</b>
    915 , &quot;A virtualization architecture for mobile WiMAX networks.&quot;
    916 SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
    917 2012.
    918 doi:10.1145/2169077.2169082.
    919 <a href="http://dx.doi.org/10.1145/2169077.2169082">http://dx.doi.org/10.1145/2169077.2169082</a>
    920 <br><br><b>Abstract: </b>Systems virtualization offers convenient means for sharing networking infrastructure while improving its utilization. This study addresses the challenges of virtualizing a commercial off-the-shelf 4G mobileWiMAX basestation. We highlight additions and modifications needed in theWiMAX network architecture for supporting multiple simultaneous virtual basestations on a single physical basestation. The most prominent features provided by the proposed virtual basestation framework include the capability to perform all frame switching at layer-2, and control mechanisms to provide isolation across slices needed to ensure experiment repeatability. By prototyping on a commercial WiMAX radio, this paper shows the usage of the virtual basestation system for housing mobile virtual network operators and testbeds alike. A use case is shown where the virtual basestation design is used to evaluate mobile handoff schemes. Another usage case is shown for optimizing a video delivery on the edge. The video delivery use case is used to show performance improvements of up to 5dB in the PSNR. Evaluation of prototype shows a significant improvement in the slice isolation, with aggregate throughput improvements of up to 192&#x0025; achievable through fair resource allocation.
    921 </li>
    922 <br>
    923 
    924 
    925 
    926 <li>
    927 <b>Blanton, Ethan and Chatterjee, Sarbajit and Gangam, Sriharsha and Kala, Sumit and Sharma, Deepti and Fahmy, Sonia and Sharma, Puneet</b>
    928 , &quot;Design and evaluation of the S<sup>3</sup> monitor network measurement service on GENI.&quot;
    929 2012 Fourth International Conference on Communication Systems and Networks (COMSNETS 2012), Bangalore, India, IEEE,
    930 2012.
    931 doi:10.1109/COMSNETS.2012.6151327.
    932 <a href="http://dx.doi.org/10.1109/COMSNETS.2012.6151327">http://dx.doi.org/10.1109/COMSNETS.2012.6151327</a>
    933 <br><br><b>Abstract: </b>Network monitoring capabilities are critical for both network operators and networked applications. In the context of an experimental test facility, network measurement is important for researchers experimenting with new network architectures and applications, as well as operators of the test facility itself. The Global Environment for Network Innovations (GENI) is a sophisticated test facility comprised of multiple ” control frameworks.” In this paper, we describe the design and implementation of S
    934 </li>
    935 <br>
    936 
    937 
    938 
    939 <li>
    940 <b>Calyam, Prasad and Venkataraman, Aishwarya and Berryman, Alex and Faerman, Marcio</b>
    941 , &quot;Experiences from Virtual Desktop CloudExperiments in GENI.&quot;
    942 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    943 2012.
    944 
    945 
    946 <br><br><b>Abstract: </b>Popular applications such as email, photo/video galleries, and file storage are increasingly being supported by cloud platforms in residential, academia and industry communities. The next frontier for these user communities will be to transition 'traditional desktops' that have dedicated hardware and software configurations into 'virtual desktop clouds' that are accessible via thin-clients. In this paper, we describe experiences from our research and development of virtual desktop cloud experiments in GENI. Our experimentation goal is to investigate and develop optimal resource allocation frameworks and performance bench- marking tools that can enable provisioning (i.e., resource sizing) and placement (i.e., resource mapping) of thin-client based virtual desktops at Internet-scale. We first motivate why virtual desktop cloud experiments cannot be done only at a table-top level, and why infrastructures such as GENI are essential. Next, we detail the methodology of our completed ” provisioning” experiments, and our work-in-progress ” placement” experiments in GENI that leverage multiple kinds of GENI resources such as aggregates, measurement services and experimenter workflow tools, as well as commercial software. Lastly, we present our vision on how our experiment slice setup and application development experiences, as well as outcomes can be leveraged in classroom labs, and 'living labs' that use GENI resources to foster training and wide- adoption of Future Internet applications.
    947 </li>
    948 <br>
    949 
    950 
    951 
    952 <li>
    953 <b>Cameron, Katherine and Brooks, R. R. and Deng, Juan and Yu, Lu and Wang, K. C. and Martin, James</b>
    954 , &quot;WiMAX: Bandwidth Contention Resolution Vulnerability to Denial of Service Attacks.&quot;
    955 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    956 2012.
    957 
    958 
    959 <br><br><b>Abstract: </b>Wireless communications is part of everyday life and 4G technology, including WiMAX, offers higher data rates and wider coverage than predecessor 3G technologies. Many security vulnerabilities have been discovered in 3G protocols and these vulnerabilities may still exist in next generation 4G protocols. This paper examines how system parameters for the WiMAX Bandwidth Contention Resolution process can affect network vulnerability to DoS attacks. It will present software simulations that explore system parameter settings and will cover the current phase of hardware simulations.
    960 </li>
    961 <br>
    962 
    963 
    964 
    965 <li>
    966 <b>Chen, Kang and Xu, Ke and Winburn, Steven and Shen, Haiying and Wang, Kuang-Ching and Li, Ze</b>
    967 , &quot;Experimentation of a MANET Routing Algorithm on the GENI ORBIT Testbed.&quot;
    968 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    969 2012.
    970 
    971 
    972 <br><br><b>Abstract: </b>This paper proposes a systematic procedure for experimentation of Mobile ad hoc networks (MANETs) on the ORBIT testbed. MANETs have attracted significant re- search interests in recent years. Most of routing or file sharing algorithms in MANETs were only evaluated by theoretical analysis or simulations because of the requirement of large scale networks. However, due to the distinctive properties of MANETs, such as mobility and decentralized structure, it has been non-trivial to deploy a real testbed for the verification. The Global Environment for Network Innovations (GENI) project sponsored by the National Science Foundation (NSF) provides an exploratory environment for academic real-world experiments, such as the ORBIT testbed. A stable and repeatable procedure for experimentation on real testbeds is necessary and important to assure the validity of results. In this paper, a MANET routing algorithm, namely LORD, was tested on the ORBIT testbed, using the proposed procedure. Specifically, we first configure the wireless interface on each node to enable the communication between each pair of nodes. Then a set of methods are adopted to construct the MANETs scenario for test. The network status is monitored throughout the entire duration of experiments. Finally, the experiment results of LORD on the GENI ORBIT testbed are demonstrated.
    973 </li>
    974 <br>
    975 
    976 
    977 
    978 <li>
    979 <b>Deng, Juan and Brooks, Richard R. and Martin, James</b>
    980 , &quot;Assessing the Effect of WiMAX System Parameter Settings on MAC-level Local DoS Vulnerability.&quot;
    981 International Journal of Performability Engineering,
    982 2012.
    983 
    984 
    985 <br><br><b>Abstract: </b>The research community has established that WiMAX networks suffer from Denial of Service (DoS) vulnerabilities. In this paper, we analyze how WiMAX system parameter settings increase or decrease DoS vulnerabilities of WiMAX networks. The behavior of the WiMAX MAC level protocol is sensitive to the settings of core system parameters. Unlike traditional network-based DoS attacks, attacks resulting from parameter misconfiguration are difficult for network operators to detect. We focus on bandwidth contention resolution aspects of the WiMAX MAC protocol. Simulations are performed using the ns-2 simulator. Analysis of Variance (ANOVA) techniques on the resulting simulation data identify which bandwidth contention resolution parameter combinations are crucial for configuring WiMAX to be less vulnerable to DoS attacks.
    986 </li>
    987 <br>
    988 
    989 
    990 
    991 <li>
    992 <b>Duerig, Jonathon and Ricci, Robert and Stoller, Leigh and Strum, Matt and Wong, Gary and Carpenter, Charles and Fei, Zongming and Griffioen, James and Nasir, Hussamuddin and Reed, Jeremy and Wu, Xiongqi</b>
    993 , &quot;Getting started with GENI: a user tutorial.&quot;
    994 SIGCOMM Comput. Commun. Rev., ACM, New York, NY, USA,
    995 2012.
    996 doi:10.1145/2096149.2096161.
    997 <a href="http://dx.doi.org/10.1145/2096149.2096161">http://dx.doi.org/10.1145/2096149.2096161</a>
    998 <br><br><b>Abstract: </b>GENI, the Global Environment for Network Innovations, is a National Science Foundation project to create a &#x76;&#x0308;irtual laboratory at the frontiers of network science and engineering for exploring future internets at scale.&#x20;&#x0308;It provides researchers, educators, and students with resources that they can use to build their own networks that span the country and - through federation - the world. GENI enables experimenters to try out bold new network architectures and designs for networked systems, and to deploy and evaluate these systems on a diverse set of resources over a large footprint. This tutorial is a starting point for running experiments on GENI. It provides an overview of GENI and covers the process of creating a network and running a simple experiment using two tools: the Flack GUI and the INSTOOLS instrumentation service.
    999 </li>
    1000 <br>
    1001 
    1002 
    1003 
    1004 <li>
    1005 <b>Duerig, Jonathon and Ricci, Robert and Stoller, Leigh and Wong, Gary and Chikkulapelly, Srikanth and Seok, Woojin</b>
    1006 , &quot;Designing a Federated Testbed as a Distributed System.&quot;
    1007 8th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TRIDENTCOM 2012),
    1008 2012.
    1009 
    1010 
    1011 <br><br><b>Abstract: </b>Traditionally, testbeds for networking and systems research have been stand-alone facilities: each is owned and operated by a single administrative entity, and is intended to be used independently of other testbeds. However, this isolated facility model is at odds with researchers' ever-increasing needs for experiments at larger scale and with a broader diversity of network technologies. The research community will be much better served by a federated model. In this model, each federated testbed maintains its own autonomy and unique strengths, but all federates work together to make their resources available under a common framework. Our challenge, then, is to design a federated testbed framework that balances competing needs: We must establish trust, but at the same time maintain the autonomy of each federated facility. While providing a unified interface to a broad set of resources, we need to expose the diversity that makes them valuable. Finally, our federation should work smoothly in a coordinated fashion, but avoid central points of failure and inter-facility dependencies. We argue that treating testbed design as a federated distributed systems problem is an effective approach to achieving this balance. The technique is illustrated through the example of ProtoGENI, a system we have designed, built, and operated according to the federated model.
    1012 </li>
    1013 <br>
    1014 
    1015 
    1016 
    1017 <li>
    1018 <b>Fund, Fraida and Dong, Chen and Korakis, Thanasis and Panwar, Shivendra</b>
    1019 , &quot;A Framework for Multidimensional Measurements on an Experimental WiMAX Testbed.&quot;
    1020 Testbeds and Research Infrastructure. Development of Networks and Communities, Springer Berlin Heidelberg,
    1021 2012.
    1022 doi:10.1007/978-3-642-35576-9&#x005F;32.
    1023 <a href="http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;32">http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;32</a>
    1024 <br><br><b>Abstract: </b>A major difficulty in the design, study, and implementation of wireless protocols and applications is the multitude of nondeterministic factors (e.g. interference, weather conditions, competing traffic) that can affect their performance. For this reason, testbeds that enable researchers to quantify these influences have become increasingly essential in the wireless research community. The growing sophistication of wireless testbeds and the wide array of services they can provide to researchers have advanced the field tremendously. Toward this end, we present an early implementation of an instrumentation and measurement framework that we have deployed on an open-access 802.16e wireless research testbed at the Polytechnic Institute of NYU. We have created a set of tools to allow experimenters to routinely collect measurements of environmental conditions during experiment runtime. These tools integrate high volumes of multidimensional measurement data from a diverse array of sources, including measurements from software defined radio peripherals, sensors, and network device drivers. With this, we aim to give researchers the ability to conduct rigorous and repeatable over-the-air experiments. We also foresee potential applications for this framework beyond its use in experiments, such as in long-term testbed monitoring.
    1025 </li>
    1026 <br>
    1027 
    1028 
    1029 
    1030 <li>
    1031 <b>Gangam, Sriharsha and Blanton, Ethan and Fahmy, Sonia</b>
    1032 , &quot;Exercises for Graduate Students using GENI.&quot;
    1033 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1034 2012.
    1035 
    1036 
    1037 <br><br><b>Abstract: </b>GENI brings together a wide variety of heterogeneous networking infrastructure and technologies under a common platform. We propose programming exercises for graduate students to introduce GENI and enable students to conduct high fidelity networking experiments. In this paper, we focus on an exercise to study congestion control and reliability using the ProtoGENI aggregate. A planned second exercise aims to leverage GENI OpenFlow aggregates to study firewalls and QoS mechanisms. We believe that these lab exercises will expose students to key networking concepts and recent research directions, e.g., in the data center context.
    1038 </li>
    1039 <br>
    1040 
    1041 
    1042 
    1043 <li>
    1044 <b>Gao, Jingcheng and Xiao, Yang</b>
    1045 , &quot;ProtoGENI DoS/DDoS Security Tests and Experiments.&quot;
    1046 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1047 2012.
    1048 
    1049 
    1050 <br><br><b>Abstract: </b>his paper will explain some tests and experiments to investigate selected security issues through ProtoGENI mainly during Spiral 3 time period and the beginning of Spiral 4. In this paper, we conduct multiple sets of DoS/ DDoS attacks in the current ProtoGENI testbed. These attacks show that it is very possible that ProtoGENI nodes may render vulnerabilities to such attacks.
    1051 </li>
    1052 <br>
    1053 
    1054 
    1055 
    1056 <li>
    1057 <b>Gember, Aaron and Dragga, Chris and Akella, Aditya</b>
    1058 , &quot;ECOS: Practical Mobile Application Offloading for Enterprises.&quot;
    1059 2nd USENIX Workshop on Hot Topics in Management of Internet, Cloud, and Enterprise Networks and Services (Hot-ICE '12),
    1060 2012.
    1061 
    1062 <a href="http://www.usenix.org/conference/hot-ice12/ecos-practical-mobile-application-of&#x0025;EF&#x0025;AC&#x0025;82oading-enterprises">http://www.usenix.org/conference/hot-ice12/ecos-practical-mobile-application-of&#x0025;EF&#x0025;AC&#x0025;82oading-enterprises</a>
    1063 <br><br><b>Abstract: </b>Offloading has emerged as a promising idea to allow handheld devices to access intensive applications without performance or energy costs. This could be particularly useful for enterprises seeking to run line-of-business applications on handhelds. However, we must address two practical roadblocks in order to make offloading amenable for enterprises: (i) ensuring data privacy and the use of trusted offloading resources, and (ii) accommodating offload at scale with diverse handheld objectives and compute resource capabilities. We present the design and implementation of an Enterprise-Centric Offloading System (ECOS) which augments prior offloading proposals to address these issues. ECOS uses a logically central controller to opportunistically leverage diverse compute resources, while tightly controlling where specific applications offload depending on privacy, performance, and energy constraints of users and applications. A wide range of experiments using a real prototype establish the effectiveness of our approach.
    1064 </li>
    1065 <br>
    1066 
    1067 
    1068 
    1069 <li>
    1070 <b>Grandl, Robert and Han, Dongsu and Lee, Suk B. and Lim, Hyeontaek and Machado, Michel and Mukerjee, Matthew and Naylor, David</b>
    1071 , &quot;Supporting network evolution and incremental deployment with XIA.&quot;
    1072 Proceedings of the ACM SIGCOMM 2012 conference on Applications, technologies, architectures, and protocols for computer communication, Helsinki, Finland, ACM, New York, NY, USA,
    1073 2012.
    1074 doi:10.1145/2342356.2342410.
    1075 <a href="http://dx.doi.org/10.1145/2342356.2342410">http://dx.doi.org/10.1145/2342356.2342410</a>
    1076 <br><br><b>Abstract: </b>eXpressive Internet Architecture (XIA) [1] is an architecture that natively supports multiple communication types and allows networks to evolve their abstractions and functionality to accommodate new styles of communication over time. XIA embeds an elegant mechanism for handling unforeseen communication types for legacy routers. In this demonstration, we show that XIA overcomes three key barriers in network evolution (outlined below) by (1) allowing end-hosts and applications to start using new communication types (e.g., service and content) before the network supports them, (2) ensuring that upgrading a subset of routers to support new functionalities immediately benefits applications, and (3) using the same mechanisms we employ for 1 and 2 to incrementally deploy XIA in IP networks.
    1077 </li>
    1078 <br>
    1079 
    1080 
    1081 
    1082 <li>
    1083 <b>Griffioen, J. and Fei, Zongming and Nasir, H. and Wu, Xiongqi and Reed, J. and Carpenter, C.</b>
    1084 , &quot;The design of an instrumentation system for federated and virtualized network testbeds.&quot;
    1085 Network Operations and Management Symposium (NOMS), 2012 IEEE, IEEE,
    1086 2012.
    1087 doi:10.1109/NOMS.2012.6212061.
    1088 <a href="http://dx.doi.org/10.1109/NOMS.2012.6212061">http://dx.doi.org/10.1109/NOMS.2012.6212061</a>
    1089 <br><br><b>Abstract: </b>Much of the GENI effort in developing network testbeds has been focused on building the control frameworks needed to allocate and initialize the network resources that make up an experiment. We argue that building the instrumentation and measurement system to monitor and capture the behavior of the network is just as important and challenging as setting up the network itself, especially in a virtualized and federated environment where getting information from experimental nodes is too complicated and too much to handle for a typical user. In this paper, we describe the design of an instrumentation and measurement infrastructure that allows users to monitor their experiments. The challenge that virtualization and federation of GENI testbeds bring to instrumentation and monitoring is how to hide the details of instrumentation setup from users so that users do not need to be experts in system administration or network management of virtualized and federated systems, but are still able to ” see” what is going on with their experiments. Our instrumentation tool sets up experiment-specific monitoring infrastructure that is tailored to capture, record, and display only information associated with that experiment. Our tools are currently available in GENI, and we present a simple example of how to use them to instrument an experiment.
    1090 </li>
    1091 <br>
    1092 
    1093 
    1094 
    1095 <li>
    1096 <b>Griffioen, James and Fei, Zongming and Nasir, Hussanmuddin and Wu, Xiongqi and Reed, Jeremy and Carpenter, Charles</b>
    1097 , &quot;Teaching with the Emerging GENI Network.&quot;
    1098 Proceedings of the 2012 International Conference on Frontiers in Education: Computer Science and Computer Engineering (FECS), Las Vegas,
    1099 2012.
    1100 
    1101 <a href="http://worldcomp-proceedings.com/proc/p2012/FEC3780.pdf">http://worldcomp-proceedings.com/proc/p2012/FEC3780.pdf</a>
    1102 <br><br><b>Abstract: </b>Over the last few years the National Science Foundation (NSF) has been investing in and developing a new network called GENI, a wide-area testbed network for at-scale experimentation with future internet designs. The GENI network has recently become available for use and is beginning to attract users. In this paper, we take a closer look at GENI with a particular focus on how GENI can be used to enhance education in the areas of computer science and computer engineering. We describe what GENI is, the resources available in GENI, and how instructors might use GENI in their classes. Being early adopters, we describe our experience using GENI in our classes, and we point out various features and challenges of using GENI. Finally, we provide tips and pointers to instructors who are interested in incorporating GENI into their own classes.
    1103 </li>
    1104 <br>
    1105 
    1106 
    1107 
    1108 <li>
    1109 <b>Huang, Shufeng and Griffioen, James and Calvert, Ken</b>
    1110 , &quot;PVNs: Making Virtualized Network Infrastructure Usable.&quot;
    1111 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS '12),
    1112 2012.
    1113 doi:10.1145/2396556.2396590.
    1114 <a href="http://dx.doi.org/10.1145/2396556.2396590">http://dx.doi.org/10.1145/2396556.2396590</a>
    1115 <br><br><b>Abstract: </b>Network virtualization is becoming a fundamental building block of future Internet architectures. Although the underlying network infrastructure needed to dynamically create and deploy custom virtual networks is rapidly taking shape ( e.g., GENI), constructing and using a virtual network is still a challenging and labor intensive task, one best left to experts. In this paper, we present the concept of a Packaged Virtual Network (PVN), that enables normal users to easily download, deploy and use application-specific virtual networks. At the heart of our approach is a PVN Hypervisor that ” runs” a PVN by allocating the virtual network resources needed by the PVN and then connecting the PVN's participants into the network on demand. To demonstrate our PVN approach, we implemented a multicast PVN that runs on the PVN hypervisor prototype using ProtoGENI as the underlying virtual network, allowing average users to create their own private multicast network.
    1116 </li>
    1117 <br>
    1118 
    1119 
    1120 
    1121 <li>
    1122 <b>Katz-Bassett, Ethan and Scott, Colin and Choffnes, David R. and Cunha, &#x49;&#x0301;talo and Valancius, Vytautas and Feamster, Nick and Madhyastha, Harsha V. and Anderson, Thomas and Krishnamurthy, Arvind</b>
    1123 , &quot;LIFEGUARD: Practical Repair of Persistent Route Failures.&quot;
    1124 Proceedings of the ACM SIGCOMM 2012 conference, ACM, New York, NY, USA,
    1125 2012.
    1126 doi:10.1145/2377677.2377756.
    1127 <a href="http://dx.doi.org/10.1145/2377677.2377756">http://dx.doi.org/10.1145/2377677.2377756</a>
    1128 <br><br><b>Abstract: </b>The Internet was designed to always find a route if there is a policy-compliant path. However, in many cases, connectivity is disrupted despite the existence of an underlying valid path. The research community has focused on short-term outages that occur during route convergence. There has been less progress on addressing avoidable long-lasting outages. Our measurements show that long-lasting events contribute significantly to overall unavailability. To address these problems, we develop LIFEGUARD, a system for automatic failure localization and remediation. LIFEGUARD uses active measurements and a historical path atlas to locate faults, even in the presence of asymmetric paths and failures. Given the ability to locate faults, we argue that the Internet protocols should allow edge ISPs to steer traffic to them around failures, without requiring the involvement of the network causing the failure. Although the Internet does not explicitly support this functionality today, we show how to approximate it using carefully crafted BGP messages. LIFEGUARD employs a set of techniques to reroute around failures with low impact on working routes. Deploying LIFEGUARD on the Internet, we find that it can effectively route traffic around an AS without causing widespread disruption.
    1129 </li>
    1130 <br>
    1131 
    1132 
    1133 
    1134 <li>
    1135 <b>Khurshid, Ahmed and Zhou, Wenxuan and Caesar, Matthew and Godfrey, P. Brighten</b>
    1136 , &quot;VeriFlow: verifying network-wide invariants in real time.&quot;
    1137 Proceedings of the first workshop on Hot topics in software defined networks, Helsinki, Finland, ACM, New York, NY, USA,
    1138 2012.
    1139 doi:10.1145/2342441.2342452.
    1140 <a href="http://doi.acm.org/10.1145/2342441.2342452">http://doi.acm.org/10.1145/2342441.2342452</a>
    1141 <br><br><b>Abstract: </b>Networks are complex and prone to bugs. Existing tools that check configuration files and data-plane state operate offline at timescales of seconds to hours, and cannot detect or prevent bugs as they arise. Is it possible to check network-wide invariants in real time, as the network state evolves? The key challenge here is to achieve extremely low latency during the checks so that network performance is not affected. In this paper, we present a preliminary design, VeriFlow, which suggests that this goal is achievable. VeriFlow is a layer between a software-defined networking controller and network devices that checks for network-wide invariant violations dynamically as each forwarding rule is inserted. Based on an implementation using a Mininet OpenFlow network and Route Views trace data, we find that VeriFlow can perform rigorous checking within hundreds of microseconds per rule insertion.
    1142 </li>
    1143 <br>
    1144 
    1145 
    1146 
    1147 <li>
    1148 <b>Kim, Hyunjun and Lee, Sungwon</b>
    1149 , &quot;FiRST Cloud Aggregate Manager development over FiRST: Future Internet testbed.&quot;
    1150 The International Conference on Information Network 2012, Bali, Indonesia, IEEE,
    1151 2012.
    1152 doi:10.1109/ICOIN.2012.6164436.
    1153 <a href="http://dx.doi.org/10.1109/ICOIN.2012.6164436">http://dx.doi.org/10.1109/ICOIN.2012.6164436</a>
    1154 <br><br><b>Abstract: </b>FiRST (Future Internet Research for Sustainable Test-bed) is the future internet platform development project being performed in Korea. The goal of the project is to create the virtualized and dynamic service creation environments over future internet networks; it is an experimental project to realize future innovative service ideas over real network environments. Among this, cloud computing is the key enabler to control and allocate virtualized network resources (such as CPU, storage, and virtualized network configuration) for the requested services. However, researches on interworking between future internet and cloud computing is in initial phase. In this paper, we propose the FiRST Cloud Aggregate Manager (AM) based on GENI (Global Environment for Network Innovation) AM Application Programming Interface (API) for the federation between future internet test-bed and open source OpenStack cloud computing platform. After that, we propose the zero-client service for mobile cloud management. In order to control the zero-client service, we develop Cloud Mobility Client/Server. And, we validate and verified our FiRST Cloud AM and zero-client service by developing experimental test-bed. Through this test-bed, we confirm that the proposed FiRST Cloud AM and zero-client service efficiently interworks with future internet control plane framework by using GENI Control Framework (GCF) tools.
    1155 </li>
    1156 <br>
    1157 
    1158 
    1159 
    1160 <li>
    1161 <b>Krishnappa, Dilip K. and Lyons, Eric and Irwin, David and Zink, Michael</b>
    1162 , &quot;Network capabilities of cloud services for a real time scientific application.&quot;
    1163 37th Annual IEEE Conference on Local Computer Networks, Clearwater Beach, FL, USA, IEEE,
    1164 2012.
    1165 doi:10.1109/lcn.2012.6423665.
    1166 <a href="http://dx.doi.org/10.1109/lcn.2012.6423665">http://dx.doi.org/10.1109/lcn.2012.6423665</a>
    1167 <br><br><b>Abstract: </b>Dedicating high-end servers for executing scientific applications that run intermittently, such as severe weather detection or generalized weather forecasting, wastes resources. While the Infrastructure-as-a-Service (IaaS) model used by today's cloud platforms is well-suited for the bursty computational demands of these applications, it is unclear if the network capabilities of today's cloud platforms are sufficient. In this paper, we analyze the networking capabilities of multiple commercial (Amazon's EC2 and Rackspace) and research (GENICloud and ExoGENI cloud) platforms in the context of a Nowcasting application, a forecasting algorithm for highly accurate, near-term, e.g., 5-20 minutes, weather predictions. The application has both computational and network requirements. While it executes rarely, whenever severe weather approaches, it benefits from an IaaS model; However, since its results are time-critical, enough bandwidth must be available to transmit radar data to cloud platforms before it becomes stale. We conduct network capacity measurements between radar sites and cloud platforms throughout the country. Our results indicate that ExoGENI cloud performs the best for both serial and parallel data transfer with an average throughput of 110.22 Mbps and 17.2 Mbps, respectively. We also found that the cloud services perform better in the distributed data transfer case, where a subset of nodes transmit data in parallel to a cloud instance. Ultimately, we conclude that commercial and research clouds are capable of providing sufficient bandwidth for our real-time Nowcasting application.
    1168 </li>
    1169 <br>
    1170 
    1171 <li>
    1172 <b>Krishnappa, Dilip K. and Lyons, Eric and Irwin, David and Zink, Michael</b>
    1173 , &quot;Performance of GENI Cloud Testbeds for Real Time Scientific Application.&quot;
    1174 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1175 2012.
    1176 
    1177 
    1178 <br><br><b>Abstract: </b>Dedicating high end servers for short-term execution of scientific applications such as weather forecasting wastes resources. Cloud platforms IaaS model seems well suited for applications which are executed on an irregular basis and for short duration. In this paper, we evaluate the performance of research testbed cloud platforms such as GENICloud and ORCA cloud clusters for our real-time scientific application of short-term weather forecasting called Nowcasting. In this paper, we evaluate the network capabilities of these research cloud testbeds for our real-time application of weather forecasting. In addition, we evaluate the computation time of executing Nowcasting on each cloud platform for weather data collected from real weather events. We also evaluate the total time taken to generate and transmit short-term forecast images to end users with live data from our own radar on campus. We also compare the performance of each of these clusters for Nowcasting with commercial cloud services such as Amazon's EC2. The results obtained from our measurement show that cloud testbeds are suitable for real-time application experiments to be carried out on a cloud platform.
    1179 </li>
    1180 <br>
    1181 
    1182 
    1183 
    1184 <li>
    1185 <b>Lee, Jae W.</b>
    1186 , &quot;Towards a Common System Architecture for Dynamically Deploying Network Services in Routers and End Hosts (Doctoral dissertation).&quot;
    1187 
    1188 2012.
    1189 
    1190 <a href="http://academiccommons.columbia.edu/download/fedora&#x005F;content/download/ac:147210/CONTENT/Lee&#x005F;columbia&#x005F;0054D&#x005F;10773.pdf">http://academiccommons.columbia.edu/download/fedora&#x005F;content/download/ac:147210/CONTENT/Lee&#x005F;columbia&#x005F;0054D&#x005F;10773.pdf</a>
    1191 <br><br><b>Abstract: </b>The architectural simplicity of the core Internet is a double-edged sword. On the one hand, its agnostic nature paved the way for endless innovations of end-to-end applications. On the other hand, the inherent limitation of this simplicity makes it difficult to add new functions to the network core itself. This is exacerbated by the conservative tendency of commercial entities to &#x6c;&#x0308;eave well-enough alone&#x2c;&#x0308; leading to the current situation often referred to as the ossification of the Internet. For decades, there has been practically no new functionality that has been added to the core Internet on a large scale. This thesis explores the possibility of enabling in-network services towards the goal of overcoming the ossification of the Internet. Our ultimate goal is to provide a common run-time environment supported by all Internet nodes and a wide-area deployment mechanism, so that network services can be freely installed, removed, and migrated among Internet nodes of all kinds–from a backbone router to a set-top box at home. In that vision of a future Internet, there is little difference between servers and routers for the purpose of running network services. Services can run anywhere on the Internet. Application service providers will have the freedom to choose the best place to run their code. This thesis presents NetServ, our first step to realize the vision of network services running anywhere on the Internet. NetServ is a node architecture for dynamically deploying in-network services on edge routers. Network functions and applications are implemented as software modules which can be deployed at any NetServ-enabled node on the Internet, subject to policy restrictions. The NetServ framework provides a common execution environment for service modules and the ability to dynamically install and remove the services without restarting the nodes. There are many challenges in designing such a system. The main contribution of this thesis lies in meeting those challenges. First, we recognize that the primary impetus for adopting new technologies is economics. To address the challenge of providing economic incentives for enabling in-network services, we demonstrate how NetServ can facilitate an economic alliance between content providers and ISPs. Using NetServ, content providers and the ISPs operating at the network edge (aka eyeball ISPs) can enter into a mutually beneficial economic relationship. ISPs make their NetServ-enabled edge routers available for hosting content providers' applications and contents. Content providers can operate closer to end users by deploying code modules on NetServ-enabled edge routers. We make our case by presenting NetServ applications which represent four concrete use cases. Second, our node architecture must support both traditional server applications and in-network packet processing applications since content providers' applications running on ISPs' routers will combine the traits of both. To address this challenge, NetServ framework can host a packet processing module that sits in the data path, a server module that uses the TCP/IP stack in the traditional way, or a combined module that does both. NetServ provides a unified runtime environment between routers and servers, taking us a step closer to the vision of the unified runtime available on all Internet nodes. Third, we must provide a fast and streamlined deployment mechanism. Content providers should be able to deploy their applications at any NetServ-enabled edge router on the Inter- net, given that they have proper authorizations. Moreover, in some application scenarios, content providers may not know the exact locations of the target routers. Content providers need a way to send a message to install or remove an application module towards a network destination, and have the NetServ-enabled routers located in the path catch and act on the message. To address this challenge, we adopted on-path signaling as the deployment mechanism for NetServ. A NetServ signaling message is sent in an IP packet towards a destination. The packet gets forwarded by IP routers as usual, but when it transits a NetServ-enabled router, the message gets intercepted and passed to the NetServ control layer. Fourth, a NetServ-enabled router must support the concurrent executions of multiple without restarting the nodes. There are many challenges in designing such a system. The main contribution of this thesis lies in meeting those challenges. First, we recognize that the primary impetus for adopting new technologies is economics. To address the challenge of providing economic incentives for enabling in-network services, we demonstrate how NetServ can facilitate an economic alliance between content providers and ISPs. Using NetServ, content providers and the ISPs operating at the network edge (aka eyeball ISPs) can enter into a mutually beneficial economic relationship. ISPs make their NetServ-enabled edge routers available for hosting content providers' applications and contents. Content providers can operate closer to end users by deploying code modules on NetServ-enabled edge routers. We make our case by presenting NetServ applications which represent four concrete use cases. Second, our node architecture must support both traditional server applications and in-network packet processing applications since content providers' applications running on ISPs' routers will combine the traits of both. To address this challenge, NetServ framework can host a packet processing module that sits in the data path, a server module that uses the TCP/IP stack in the traditional way, or a combined module that does both. NetServ provides a unified runtime environment between routers and servers, taking us a step closer to the vision of the unified runtime available on all Internet nodes. Third, we must provide a fast and streamlined deployment mechanism. Content providers should be able to deploy their applications at any NetServ-enabled edge router on the Internet, given that they have proper authorizations. Moreover, in some application scenarios, content providers may not know the exact locations of the target routers. Content providers need a way to send a message to install or remove an application module towards a network destination, and have the NetServ-enabled routers located in the path catch and act on the message. To address this challenge, we adopted on-path signaling as the deployment mechanism for NetServ. A NetServ signaling message is sent in an IP packet towards a destination. The packet gets forwarded by IP routers as usual, but when it transits a NetServ-enabled router, the message gets intercepted and passed to the NetServ control layer. Fourth, a NetServ-enabled router must support the concurrent executions of multiple content providers' applications. Each content provider's execution environment must be isolated from one another, and the resource usage of each must be controlled. To address the challenge of providing a robust multi-user execution environment, we chose to run NetServ modules in user space. This is in stark contrast to most programmable routers, which run service modules in kernel space for fast packet processing. Furthermore, NetServ modules are written in Java and run in Java Virtual Machines (JVMs). Our choice of user space execution and JVM allows us to leverage the decades of technology advances in operating systems, virtualization, and Java. Lastly, in order to host the services of a large number of content providers, NetServ must be able to scale beyond the single-box architecture. We address this challenge with the multi-box lateral expansion of NetServ using the OpenFlow forwarding engine. In this extended architecture, multiple NetServ nodes are attached to an OpenFlow switch, which provides a physically separate forwarding plane. The scalability of user services is no longer limited to a single NetServ box. Additionally, this thesis presents our prior work on improving service discovery in local and global networks. The service discovery work makes indirect contribution because the limitations of local and overlay networks encountered during those studies eventually led us to investigate in-network services, which resulted in NetServ. Specifically, we investigate the issues involved in bootstrapping large-scale structured overlay networks, present a tool to merge service announcements from multiple local networks, and propose an enhancement to structured overlay networks using link-local multicast.
    1192 </li>
    1193 <br>
    1194 
    1195 
    1196 
    1197 <li>
    1198 <b>Li, Ting and Van Vorst, Nathanael and Rong, Rong and Liu, Jason</b>
    1199 , &quot;Simulation studies of OpenFlow-based in-network caching strategies.&quot;
    1200 Proceedings of the 15th Communications and Networking Simulation Symposium, Orlando, Florida, Society for Computer Simulation International, San Diego, CA, USA,
    1201 2012.
    1202 
    1203 <a href="http://portal.acm.org/citation.cfm?id=2331762.2331774">http://portal.acm.org/citation.cfm?id=2331762.2331774</a>
    1204 <br><br><b>Abstract: </b>We propose an in-network caching architecture using Open-Flow to coordinate caching decisions in the network. Our scheme, called CacheFlow, extends the cache-and-forward concept by moving contents closer to the clients hop-by-hop using TCP for sending requests and retrieving contents. As such, CacheFlow can be incrementally implemented and deployed in the real network. In this paper, we present a simulation study of several caching policies, including a random cache policy, a statically optimal cache placement policy and a new disk placement strategy that places popular contents at the &#x63;&#x0308;enter&#x20;&#x0308;of the network. Experimental results show that simple in-network caching policies can be realized using today's technology to improve network performance.
    1205 </li>
    1206 <br>
    1207 
    1208 
    1209 
    1210 <li>
    1211 <b>Liu, Jun and O'Neil, Thomas and Desell, Travis and Carlson, Ross</b>
    1212 , &quot;Work-in-Progress: Empirical Verification of A Subset Sum Hypothesis in GENI Cloud.&quot;
    1213 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1214 2012.
    1215 
    1216 
    1217 
    1218 </li>
    1219 <br>
    1220 
    1221 
    1222 
    1223 <li>
    1224 <b>Luna, Nicholas and Shetty, Sachin and Rogers, Tamara and Xiong, Kaiqi</b>
    1225 , &quot;Assessment of Router Vulnerabilities on PlanetLab Infrastructure for Secure Cloud Computing.&quot;
    1226 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1227 2012.
    1228 
    1229 
    1230 <br><br><b>Abstract: </b>In recent times, the cloud computing based delivery model has been proven to reduce enterprise IT costs and complexities. In contrast to traditional enterprise IT solutions, the cloud computing model moves the application software and data to remote servers in large datacenters, which raises many security challenges. One of the critical challenges is the inability to characterize the impact of the vulnerabilities of routers on the cloud security and performance guarantees. In this paper, we analyze the degree of security provided by routers to data sharing applications deployed in cloud environments that span administrative and network domains. Our analysis is based on examining the security level of network applications on routers which lie between nodes on Planetlab infrastructure. We assume that some of the PlanetLab nodes will share the same wide area network path as the cloud servers. Our preliminary results confirm that the majority of the routers are plagued by insecure network protocols, leading to vulnerable routers. These results confirm our hypothesis that the security of the network infrastructure needs to be upgraded to assure the protection of information exchanged on the wide area network path.
    1231 </li>
    1232 <br>
    1233 
    1234 
    1235 
    1236 <li>
    1237 <b>Maccherani, E. and Femminella, M. and Lee, J. W. and Francescangeli, R. and Janak, J. and Reali, G. and Schulzrinne, H.</b>
    1238 , &quot;Extending the NetServ autonomic management capabilities using OpenFlow.&quot;
    1239 2012 IEEE Network Operations and Management Symposium, Maui, HI, IEEE,
    1240 2012.
    1241 doi:10.1109/NOMS.2012.6211961.
    1242 <a href="http://dx.doi.org/10.1109/NOMS.2012.6211961">http://dx.doi.org/10.1109/NOMS.2012.6211961</a>
    1243 <br><br><b>Abstract: </b>Autonomic management capabilities of the Future Internet can be provided through a recently proposed service architecture called NetServ. It consists of the interconnection of programmable nodes which enable dynamic deployment and execution of network and application services. This paper shows how this architecture can be further improved by introducing the OpenFlow architecture and implementing the OpenFlow controller as a NetServ service, thus improving both the NetServ management performance and its flexibility. These achievements are demonstrated experimentally on the GENI environment, showing the platform self-protecting capabilities in case of a SIP DoS attack.
    1244 </li>
    1245 <br>
    1246 
    1247 
    1248 
    1249 <li>
    1250 <b>Mandvekar, Lokesh and Sathyaraja, Anandatirtha and Qiao, Chunming</b>
    1251 , &quot;Socially Aware Single System Images.&quot;
    1252 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1253 2012.
    1254 
    1255 
    1256 <br><br><b>Abstract: </b>Cloud computing enables users to get access to huge amounts of computing resources as desired. There are many popular commercial cloud service providers which provide resources to users at a price. These providers can not be trusted as far as privacy of data is concerned. On the other hand, people do trust their close friends, relatives and other social contacts, albeit, to varying degrees. This paper reports the work-in-progress on S3I(Socially Aware Single System Images) which allows users to form computing clusters using resources owned by their social contacts. It tries to utilize the trust found between people in real life and translate it to provide trustworthy resource sharing between them.
    1257 </li>
    1258 <br>
    1259 
    1260 
    1261 
    1262 <li>
    1263 <b>Maziku, Hellen and Shetty, Sachin and Rogers, Tamara</b>
    1264 , &quot;Measurement-based IP Geolocation of Routers on Planetlab Infrastructure.&quot;
    1265 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1266 2012.
    1267 
    1268 
    1269 <br><br><b>Abstract: </b>Location aware applications can benefit from a more accurate yet robust IP geolocation framework. Various approaches to IP geolocation have been well documented. The most recent approach casts IP geolocation as a machine learn- ing classification problem. This approach makes it possible to incorporate both delay and non delay based information. The accuracy of IP geolocation can be improved by incorporating additional types of geolocation information rather relying on network delay alone. To enhance the classification accuracy of the existing classification framework, we expand it to include 6 features (3 of which are novel). We use PlanetLab as a testbed to generate our measurement set. We select 67 PlanetLab nodes within the United States with known geographic location as our landmarks. We test the accuracy of our framework on 23,843 routers given ping measurements from the 67 landmarks. With only three features (average delay, average hops and population density) tested, our new classifier gives a reduced average error distance of 157.81 miles and a median error distance of 0 miles, compared to the present classifier that gives an average error distance of 253.34 miles. This is very promising as we move on to the next phase of incorporating data for the remaining 5 features. To the best of our knowledge, this is the first proposed framework that aims to improve the accuracy of the present classifier based IP geolocation.
    1270 </li>
    1271 <br>
    1272 
    1273 
    1274 
    1275 <li>
    1276 <b>Mitroff, Sarah</b>
    1277 , &quot;Lawrence Landweber Helped Build Today's Internet, Now He's Advising Its Future.&quot;
    1278 Wired,
    1279 2012.
    1280 
    1281 <a href="http://www.wired.com/business/2012/08/lawrence-landweber/">http://www.wired.com/business/2012/08/lawrence-landweber/</a>
    1282 
    1283 </li>
    1284 <br>
    1285 
    1286 
    1287 
    1288 <li>
    1289 <b>Muhammad, Monzur and Cappos, Justin</b>
    1290 , &quot;Towards a Representive Testbed: Harnessing Volunteers for Networks Research.&quot;
    1291 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1292 2012.
    1293 
    1294 
    1295 <br><br><b>Abstract: </b>A steady rise in home systems has been seen over the past few years. As more systems are designed and deployed, an appropriate testbed is required to test these systems. Sev- eral systems exist, such as PlanetLab, that currently provide a networking testbed allowing researchers and developers to test and measure various applications. However in the long run such testbeds will be unable to keep up and meet all the demands of many of the large scale modern day peer-to-peer systems. We outline the various challenges and essentials of a networking testbed and we provide an alternate network- ing testbed that is driven by resources that are voluntarily contributed. We talk about the various advantages and dis- advantages of the Seattle system, an open source peer-to- peer computing testbed that has the potential to meet these demands. The testbed is composed of sandboxed resources that are donated by volunteers. Seattle has been deployed for about three years and supports many researchers who are interested in a networking testbed. The testbed consists of over 4100 nodes and is constantly growing. Seattle looks to grow and meet the demands of networking testbeds as they are made.
    1296 </li>
    1297 <br>
    1298 
    1299 
    1300 
    1301 <li>
    1302 <b>Ozcelik, Ilker and Brooks, Richard R.</b>
    1303 , &quot;Performance Analysis of DDoS Detection Methods on Real Network.&quot;
    1304 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1305 2012.
    1306 
    1307 
    1308 <br><br><b>Abstract: </b>Distributed Denial of Service (DDoS) attacks are major security threats to the Internet. The distributed structure of these attacks makes it difficult to distinguish between legitimate and attack traffic, making detection difficult. In addition to this challenge, researchers also have to study and find countermeasures against these attacks without using an operational network for testing, since attacks on operational networks inconvenience users. In this paper, we propose a method to perform DDoS analysis on real hardware using real traffic without jeopardizing the original network. We implement our experiments on the Geni testbed using Openflow. We present results from DDoS detection methods using operational traffic.
    1309 </li>
    1310 <br>
    1311 
    1312 
    1313 
    1314 <li>
    1315 <b>Qin, Z. and Xiong, X. and Chuah, M.</b>
    1316 , &quot;Lehigh Explorer: Android Application Utilizing Content Centric Features.&quot;
    1317 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1318 2012.
    1319 
    1320 
    1321 <br><br><b>Abstract: </b>Companies, government organizations or institutions from anywhere in the world publish different types of information e.g. news, health alerts, disaster warnings at any time. Rather than consuming all published data, users only desire access to information of interest to themselves irrespective of where the data is located and who publish them. Existing publish/subscribe systems built based on IP-based network can be inefficient and are not flexible enough to meet emerging requirements e.g. deal with mobile users, dynamic contents, searching over encrypted data. Recently content-centric networks have been proposed to provide flexibility to users to access such information. We have designed secure content centric mobile networks that allow users to publish and retrieve contents securely. As with any new architecture, one important issue is to have useful applications that can utilize features provided in the new architecture. In this paper, we describe an Android application we recently developed that allows visitors to explore Lehigh campus based on their expressed interests. Our application utilizes keyword based interest messages to retrieve matching data items of interests to a user. We are giving a demo of Lehigh Explorer at GEC13.
    1322 </li>
    1323 <br>
    1324 
    1325 
    1326 
    1327 <li>
    1328 <b>Raychaudhuri, Dipankar and Nagaraja, Kiran and Venkataramani, Arun</b>
    1329 , &quot;MobilityFirst: a robust and trustworthy mobility-centric architecture for the future internet.&quot;
    1330 SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
    1331 2012.
    1332 doi:10.1145/2412096.2412098.
    1333 <a href="http://dx.doi.org/10.1145/2412096.2412098">http://dx.doi.org/10.1145/2412096.2412098</a>
    1334 <br><br><b>Abstract: </b>This paper presents an overview of the MobilityFirst network architecture, currently under development as part of the US National Science Foundation's Future Internet Architecture (FIA) program. The proposed architecture is intended to directly address the challenges of wireless access and mobility at scale, while also providing new services needed for emerging mobile Internet application scenarios. After briefly outlining the original design goals of the project, we provide a discussion of the main architectural concepts behind the network design, identifying key features such as separation of names from addresses, public-key based globally unique identifiers (GUIDs) for named objects, global name resolution service (GNRS) for dynamic binding of names to addresses, storage-aware routing and late binding, content- and context-aware services, optional in-network compute layer, and so on. This is followed by a brief description of the MobilityFirst protocol stack as a whole, along with an explanation of how the protocol works at end-user devices and inside network routers. Example of specific advanced services supported by the protocol stack, including multi-homing, mobility with disconnection, and content retrieval/caching are given for illustration. Further design details of two key protocol components, the GNRS name resolution service and the GSTAR routing protocol, are also described along with sample results from evaluation. In conclusion, a brief description of an ongoing multi-site experimental proof-of-concept deployment of the MobilityFirst protocol stack on the GENI testbed is provided.
    1335 </li>
    1336 <br>
    1337 
    1338 
    1339 
    1340 <li>
    1341 <b>Rosen, Aaron</b>
    1342 , &quot;Network Service Delivery and Throughput Optimization via Software Defined Networking (Master's Thesis).&quot;
    1343 
    1344 2012.
    1345 
    1346 <a href="http://tigerprints.clemson.edu/all&#x005F;theses/1332/">http://tigerprints.clemson.edu/all&#x005F;theses/1332/</a>
    1347 <br><br><b>Abstract: </b>In today's world, transmitting data across large bandwidth-delay product (BDP) networks requires special configuration on end users' machines in order to be done efficiently. This added level of complexity creates extra cost and is usually overlooked by users unknowledgeable to the issues. This is one example problem which can be ameliorated with the emerging software defined networking (SDN) paradigm. In an SDN, packet forwarding is controlled via software controllers. In an OpenFlow SDN, a controller can control the forwarding, rewriting, and dropping of packets based on their header attributes. The ability to handle packets in customizable ways in software has significant implications for both users and operators of the network. Via SDN, network providers can easily provide services to enhance users' experience of the network. Steroid OpenFlow Service (SOS) is presented as a solution to seamless enhancement of TCP data transfer throughput over large BDP networks without any modification to the software and configurations on users' machines. SOS utilizes OpenFlow to redirect application specific traffic to application specific service agents. SOS uses service agents on both ends of the connection to seamlessly terminate a user's TCP connection, launch a set of parallel TCP connections, and leverage multiple paths when available to maximize throughput.
    1348 </li>
    1349 <br>
    1350 
    1351 
    1352 
    1353 <li>
    1354 <b>Rosen, Aaron and Wang, Kuang-Ching</b>
    1355 , &quot;Steroid OpenFlow Service: Seamless Network Service Delivery in Software Defined Networks.&quot;
    1356 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1357 2012.
    1358 
    1359 
    1360 <br><br><b>Abstract: </b>In a software defined network (SDN), packet forwarding is controlled by software controllers. In an OpenFlow SDN, a controller can control the forwarding, rewriting, and dropping of packets based on their header attributes. The ability to handle packets in customizable ways in software has significant implications for both network users and operators. Via software, users can convey application specific expectations while operators can deliver application specific services to enhance user experiences. In this paper, we present the Steroid OpenFlow Services (SOS) paradigm for network services delivery. The paradigm enables operators to deliver network services without any setup requirements on user machines. SOS utilizes OpenFlow to redirect application specific traffic to application specific service agents; SOS also rewrites packet headers for a service to remain seamless to users. This paper presents an example SOS service for optimizing large volume TCP download across a large delay-bandwidth-product wide area network. SOS service agents on both ends of the connection seamlessly terminate a user TCP connection, launch a set of parallel TCP connections, and leverage multiple paths when available to maximize throughput. With the NSF GENI future Internet testbed, a prototype implementation achieved up to 320 times throughput enhancement seamless to the end users.
    1361 </li>
    1362 <br>
    1363 
    1364 
    1365 
    1366 <li>
    1367 <b>Shin, Sunae and Dhondge, Kaustubh and Choi, Baek-Young</b>
    1368 , &quot;Understanding the Performance of TCP and UDP-based Data Transfer Protocols using EMULAB.&quot;
    1369 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1370 2012.
    1371 
    1372 
    1373 <br><br><b>Abstract: </b>In this paper, we present a hands-on course project that explores the performance of data transfer protocols using a GENI resource. TCP is one of the key topics in networking courses, and understanding its behavior as well as limitations, from real experiments, offers an invaluable and deep learning experience. A protocol's performance is directly impacted by network parameters such as network bandwidth, delay and loss. However, it is difficult to control and even vary those parameters, if it is not evaluated with simulations. GENI facilities conveniently provide a virtual laboratory that enables us to control the network settings with real network systems. Through this educational project, students had an opportunity to control important network parameters, and measure and compare TCP's performance with a UDP-based data transfer protocol, UDT, using EMULAB. Students were enthusiastic to witness the protocols' performances, and the limitations of TCP under a high bandwidth delay product network in the presence of packet loss, and to recognize the importance of protocol design and system issues for the future Internet.
    1374 </li>
    1375 <br>
    1376 
    1377 
    1378 
    1379 <li>
    1380 <b>Sivakumar, Ashiwan and Shankaranarayanan, P. N. and Rao, Sanjay</b>
    1381 , &quot;Closer to the Cloud - A Case for Emulating Cloud Dynamics by Controlling the Environment.&quot;
    1382 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1383 2012.
    1384 
    1385 
    1386 
    1387 </li>
    1388 <br>
    1389 
    1390 
    1391 
    1392 <li>
    1393 <b>Soroush, Hamed and Banerjee, Nilanjan and Corner, Mark and Levine, Brian and Lynn, Brian</b>
    1394 , &quot;A retrospective look at the UMass DOME mobile testbed.&quot;
    1395 SIGMOBILE Mob. Comput. Commun. Rev., ACM, New York, NY, USA,
    1396 2012.
    1397 doi:10.1145/2169077.2169079.
    1398 <a href="http://dx.doi.org/10.1145/2169077.2169079">http://dx.doi.org/10.1145/2169077.2169079</a>
    1399 <br><br><b>Abstract: </b>In this paper we describe the evolution of DOME, a diverse outdoor testbed for mobile experimentation. In addition, while highlighting the challenges faced in construction of DOME, we describe a concrete set of scientific results derived from this experience in a retrospective study. First, we argue that a broad range of mobility experiments could be performed in a testbed which provides the properties of temporal, technological, and spatial diversity. We demonstrate these properties in our testbed through analysis of data collected from DOME over a period of four years. Second, we crystallize a set of design principles that others should use when constructing testbeds of their own, including those related to deploying and managing a diverse testbed, distributing experiments remotely, and fostering collaborations among testbed stakeholders. Finally, using traces collected by DOME, we provide insights into several important problems in mobile systems research.
    1400 </li>
    1401 <br>
    1402 
    1403 
    1404 
    1405 <li>
    1406 <b>Stabler, Greg and Goasguen, Sebastien and Rosen, Aaron and Wang, Kuang-Ching</b>
    1407 , &quot;OneCloud: Controlling the Network in an OpenFlow Cloud.&quot;
    1408 First GENI Research and Educational Experiment Workshop (GREE 2012), Los Angeles,
    1409 2012.
    1410 
    1411 
    1412 <br><br><b>Abstract: </b>Cloud computing is an emerging paradigm for on-demand access to computing resources over the network. Beyond early Software as a Service (SaaS) offerings, there is an increasing interest in the Infrastructure as a Service (IaaS) model where users request specific storage, networking, and computing resources to meet their application needs. To provision the network in a cloud, IaaS providers, such as the Amazon Web Services, allow users to choose their IP addresses, which can be associated with a dynamic set of virtual hosts (Elastic IP) with VPN, dynamic DNS, and dynamic firewall services. In this paper, we analyze a range of cloud network provisioning needs and the means to realize them in an OpenFlow network. We present an OpenFlow enabled framework for cloud network provisioning, based on the Open- Nebula cloud provisioning engine. Specifically, we demonstrate an Elastic IP service compatible with the Amazon Elastic Compute Cloud (EC2) API. This demonstration is available on the Clemson OneCloud IaaS offering. Ongoing efforts focus on the enablement of additional cloud network services for campus networks and wide area experimental networks like the National Science Foundation's GENI network.
    1413 </li>
    1414 <br>
    1415 
    1416 
    1417 
    1418 <li>
    1419 <b>Stabler, Greg and Rosen, Aaron and Goasguen, Sebastien and Wang, Kuang-Ching</b>
    1420 , &quot;Elastic IP and security groups implementation using OpenFlow.&quot;
    1421 Proceedings of the 6th international workshop on Virtualization Technologies in Distributed Computing Date, Delft, The Netherlands, ACM, New York, NY, USA,
    1422 2012.
    1423 doi:10.1145/2287056.2287069.
    1424 <a href="http://doi.acm.org/10.1145/2287056.2287069">http://doi.acm.org/10.1145/2287056.2287069</a>
    1425 <br><br><b>Abstract: </b>This paper presents a reference implementation of an Elastic IP and Security Group service using the OpenFlow protocol. The implementation is the first to present integration of OpenFlow within a virtual machine provisioning engine and an API for enabling such services. In this paper the OpenNebula system is used. The Elastic IP and Security Groups services are similar to the Amazon EC2 services and present a compatible Query API implemented by OpenNebula. The core of the implementation relies on the integration of an OpenFlow controller (NOX) with the EC2 server. Flow rules can be inserted in the OpenFlow controller using the EC2 API. These rules are then used by Open vSwitch bridges on the underlying hypervisor to manage network traffic. The reference implementation presented opens the door for more advanced cloud networking services that leverage principles from software defined networking including virtual private cloud, virtual data center spanning multiple availability zones, as well as seamless migration over wide are networks.
    1426 </li>
    1427 <br>
    1428 
    1429 
    1430 
    1431 <li>
    1432 <b>Teerapittayanon, Surat and Fouli, Kerim and M&#x65;&#x0301;dard, Muriel and Montpetit, Marie-Jos&#x65;&#x0301; and Shi, Xiaomeng and Seskar, Ivan and Gosain, Abhimanyu</b>
    1433 , &quot;Network Coding as a WiMAX Link Reliability Mechanism.&quot;
    1434 Multiple Access Communications, Springer Berlin Heidelberg,
    1435 2012.
    1436 doi:10.1007/978-3-642-34976-8&#x005F;1.
    1437 <a href="http://dx.doi.org/10.1007/978-3-642-34976-8&#x005F;1">http://dx.doi.org/10.1007/978-3-642-34976-8&#x005F;1</a>
    1438 <br><br><b>Abstract: </b>We design and implement a network-coding-enabled relia- bility architecture for next generation wireless networks. Our network coding (NC) architecture uses a flexible thread-based design, with each encoder-decoder instance applying systematic intra-session random lin- ear network coding as a packet erasure code at the IP layer. Using GENI WiMAX platforms, a series of point-to-point transmission experiments were conducted to compare the performance of the NC architecture to that of the Automatic Repeated reQuest (ARQ) and Hybrid ARQ (HARQ) mechanisms. In our scenarios, the proposed architecture is able to decrease packet loss from around 11-32&#x0025; to nearly 0&#x0025;; compared to HARQ and joint HARQ/ARQ mechanisms, the NC architecture offers up to 5.9 times gain in throughput and 5.5 times reduction in end-to- end file transfer delay. By establishing NC as a potential substitute for HARQ/ARQ, our experiments offer important insights into cross-layer designs of next generation wireless networks.
    1439 </li>
    1440 <br>
    1441 
    1442 
    1443 
    1444 <li>
    1445 <b>Thomas, Charles and Sommers, Joel and Barford, Paul and Kim, Dongchan and Das, Ananya and Segebre, Roberto and Crovella, Mark</b>
    1446 , &quot;A Passive Measurement System for Network Testbeds.&quot;
    1447 Testbeds and Research Infrastructure. Development of Networks and Communities, Springer Berlin Heidelberg,
    1448 2012.
    1449 doi:10.1007/978-3-642-35576-9&#x005F;14.
    1450 <a href="http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;14">http://dx.doi.org/10.1007/978-3-642-35576-9&#x005F;14</a>
    1451 <br><br><b>Abstract: </b>The ability to capture and process packet-level data is of intrinsic importance in network testbeds that offer broad experimental capabilities to researchers. In this paper we describe the design and implementation of a passive measurement system for network testbeds called GIMS. The system enables users to specify and centrally manage packet capture on a set of dedicated measurement nodes deployed on links in a distributed testbed. The first component of GIMS is a scalable experiment management system that coordinates multi-tenant access to measurement nodes through a web-based user interface. The second component of GIMS is a node management system that enables (i) local processing on packets (e.g., flow aggregation and sampling), (ii) meta-data to be added to captured packets (e.g., timestamps), (iii) packet anonymization per local security policy, and (iv) flexible data storage including transfer to remote archives. We demonstrate the capabilities of GIMS through a set of micro-benchmarks that specifically highlight t