GEC 11 Poster Descriptions
- A New Generation Network Architecture to Accommodate Virtual Network …
- Applying a Distributed Security Sensor Network to GENI (HiveMind)
- DCN Deployment on JGN-X and Collaboration with Network Virtualization …
- DDoS Attack Detection & DoS Attacks Exploiting WiMAX System Parameters
- Developing GENI Aggregates for Real-TIme Large-Scale Network Simulation …
- Evaluating Schemes for Adapting to Cloud Dynamics using GENI
- GENI Experiments on P2P and MANET Networks
- GENI Meta Operation Center (GMOC)
- GENI Monitoring Slice: Enabling Network Visibility in the GENI OpenFlow …
- GENI Research at Georgia Tech (BGPMux)
- GENI-fying Federated, Autonomous Wireless Sensor Network Infrastructures …
- Great Plains Environment for Network Innovation (GpENI)
- GridStat on GENI : Simulating a Smart Power Grid Infrastructure over GENI …
- Infinity: In-Network Storage for Mobile Devices
- Integration for GENI Experiments and Measurement Archive (OnTimeMeasure)
- Integration of LEARN with GENI Infrastructures using ORCA: VLAN …
- Leveraging and Abstracting Measurements with PerfSONAR (LAMP)
- MAX GENI Aggregate Federation and Stitching
- Measurement Data Archive (DigitalObjectRegistry)
- Network Virtualization Research In Japan
- OpenFlow Campus Trial at Clemson: An OpenFlow Service for Seamless …
- Scalable Sensing Service (S3MONITOR)
- Secure Content Centric Mobile Network (SECON)
- Socially Aware Single System Image
- SPP Deployment and Named Data Networking Research
- Taiwan Integrated Research Network (iGENI)
- The Performance Evaluation of Bandwidth Allocation Algorithms in …
- TransCloud (GENICloud)
- Trema; An Open Source OpenFlow Controller Platform
- TUNIE: A Flexible and Programmable Virtualized Network Innovation …
- Using OMF/OML for WiMAX Experiments (WIMXBBN)
- Wide-area programmable network: Operation And Management of a nation wide …
A New Generation Network Architecture to Accommodate Virtual Network Application Service Providers
Eiji Kawai, National Institute of Information and Communications Technology (NICT), Japan
Shuji Ishii, National Institute of Information and Communications Technology (NICT), Japan
Hiroaki Yamanaka, National Institute of Information and Communications Technology (NICT), Japan
Katsuyoshi Iida, Tokyo Institute of Technology, Japan
Masayoshi Shimamura, Tokyo Institute of Technology, Japan
Takuya Omizo, Tokyo Institute of Technology, Japan
Masato Tsuru, Kyushu Institute of Technology, Japan
An essential issue in the future Internet is how to efficiently manage the network and computational resources shared by a variety of application services with different QoS requirements over multiple diverse infrastructural networks. Network virtualization is a promising approach but further studies to develop an effective and practical architecture are required. Our poster presents an outline of a proposed architecture aiming to efficiently accommodate heterogeneous and numerous virtual network application service providers assuming the use of OpenFlow-based technology. In particular, the architecture focuses on the following features: (1) OpenFlow network virtualization in a large-scale testbed environment, (2) management of resources by a meta-resource provider to accommodate diverse QoS requirements and dynamic resource status over infrastructural network domains, and (3) distributed information management of network and computational resources based on perfSONAR technology.
- Future Internet
- New generation network architecture
- Network Virtualization
Applying a Distributed Security Sensor Network to GENI (HiveMind)
Sean Peisert, University of California, Davis (PI)
Matt Bishop, University of California, Davis
Steven Templeton, University of California, Davis
Carrie Gates, CA Labs (CoPI)
Securing GENI is a challenging task. Its highly distributed and autonomous infrastructure and a requirement that security monitoring and response must notadversely affect networking experiments running on GENI slices both define the problem and limit solutions. Three primary challenges must be managed. First,because the GENI infrastructure is provided by many independent Aggregators,security policy can vary widely; each Aggregator is free to determine what activity is allowed and how violations will be handled. As a result, no standardsecurity policy can be enforced across an entire distributed GENI slice. Second,because of its distributed infrastructure, a centralized IDS is not feasible.Ultimately, monitoring must be done locally, in each Aggregator and on each node -- no central point for monitoring network traffic exists, and in addition to local policy issues, the bandwidth required to forward all collected network and host data to a central point would be excessive and affects the third issue, harming GENI experiments. Security monitoring solutions that use significantamounts of network or host resources may disrupt GENI networking experiments, particularly those with strict timing requirements. This is a problemwith both network and host based monitoring, particularly if the monitored devices are resource constrained such as cyber-physical devices or when highlyvirtualized. Traditional IDS methods relying on continuous full monitoring of all hosts and network activity are not suitable.Our project -- the Hive Mind -- investigates a biologically-inspired, scalable, lightweight, decentralized security event monitoring alternative to "running all sensors, at all times, on all nodes."Instead, using behavior of social insects such as ants, bees and wasps, and higher animals such as crows and wolves, we relyon emergent behavior to determine which sensors to run when and where. Virtual creatures move across the mesh of nodes initiating sensor functions andcommunicating between each other. This dynamically and without external direction, focuses monitoring to areas of immediate interest while minimizingresource overhead. Additionally, by providing directed, limited cross-node communication of security relevant events, we can achieve the equivalent ofcentralized monitoring, without the degradation of resources available to experiments. For example, detecting attacks launched from GENI experiments are a specificconcern of the GENI Project Office. Because these may be distributed attacks,not detectible from a single host or Aggregator, they pose a significant challenge. One such attack would be a distributed denial of service attack (DDoS)launched from a large GENI experiment against an Internet host. The activity ofindividual nodes, in isolation would not be suspicious. Only in aggregate, seenacross the nodes in the experiment, can the activity be considered an attack.Other attacks, such as those directed to a single node, are detected locally, butthe suspicious activity is communicated to neighboring nodes to quickly informothers of the problem so that they may determine if they also were targeted.This poster provides an overview of our research and implementation of thismodel for security event monitoring for GENI.
DCN Deployment on JGN-X and Collaboration with Network Virtualization Projects
Hideki Otsuki, NICT
We introduce our Dynamic Circuit Network deployment on our new JGN-X testbed. Previously,we were deployed DCN-SS and EX4200 are used for DCN-switch. These switches are located at Ingress and Egress point. In JGN-X, MX-80 is introduced as core routers and DCN network will created on a virtual router on MX-80. DCN is possible to deploy more widely in Japan. And we will continue to collaborate with CoreLab project. PLC of CoreLab will make request VLAN path among slices to DCN via DCN-API.
DDoS Attack Detection & DoS Attacks Exploiting WiMAX System Parameters
Ilker Ozcelik, Holcombe Department of Electrical & Computer Engineering
Lu Yu, Clemson University
The poster comprises two parts. In part one, we are collecting the Internet traffic signature on OpenFlow to use as backgroutnd traffic. By using the real background traffic, we are investigating the effectiveness of theoretical DDoS attack detection techniques on GENI. We are also trying to evaluate our proposed equation of necessary traffic for DDoS attack. Part two focuses on analyzing DoS attacks that exploit WiMAX system parameter settings. We concentrate on parameters concerning bandwidth contention resolution in IEEE 802.16 standards. We use analysis of variance (ANOVA) to find how parameter settings affect the ability of DoS attackers to monopolize network bandwidth. We are carrying out a DoS attack against WiMAX on GENI ORBIT and collecting the data used for ANOVA.
Developing GENI Aggregates for Real-TIme Large-Scale Network Simulation (PrimoGENI)
Nathanael Van Vorst, School of Computing and Information Sciences, Florida International University
Miguel Erazo, School of Computing and Information Sciences, Florida International University
Hao Jiang, School of Computing and Information Sciences, Florida International University
Ting Li, School of Computing and Information Sciences, Florida International University
Jason Liu, School of Computing and Information Sciences, Florida International University
The goal of PrimoGENI is to incorporate real-time network simulation into the GENI "ecosystem". We have extended PRIME, our existing real-time large-scale network simulator, to become part of the GENI federation. PrimoGENI will support large-scale GENI experiments with millions of simulated network entities (hosts, routers, and links) and thousands of emulated elements running unmodified network protocols and applications.
Evaluating Schemes for Adapting to Cloud Dynamics using GENI
Ashiwan Sivakumar, Purdue University
Shankaranarayanan PN, Purdue University
Mohammad Hajjat, Purdue University
Dr. Sanjay Rao, Purdue University
Enterprises are increasingly deploying their applications in the cloud given the cost-saving advantages, and the potential to geo-distribute applications to ensure resilience and better service experience. Latency and availability are critical with such performance sensitive applications. A key problem then is to meet the stringent response time requirements of enterprise applications in the cloud. We build a system that we term Dealer which for each component, dynamically splits transactions among its replicas in different data-centers. It adapts to sudden changes in delay across components and routes requests to replicas of the components in a different data-center. In doing so, Dealer seeks to minimize user response times, and takes component performance, as well as intra-data- center and inter-data-center communication latencies into account. We have integrated the system with a performance sensitive trading application called Daytrader in GENI.
Our approach to evaluate the system makes use of the controlled and repeatable environment provided by GENI. The experiments that we have conducted on GENI aim at emulating sudden spikes in delay between components. We have studied the dynamic response time of Dealer by subjecting it to a Step Up input reference waveform. We have also compared the user response times in a Multi cloud environment on ProtoGENI with and without dealer. We present the results of the evaluation experiments conducted on GENI.
GENI Experiments on P2P and MANET Networks
Haiyin (Helen) Shen, Clemson University
Kuang-Ching Wang, Clemson University
Kang Chen, Clemson University
Ke Xu, Clemson University
Steven Winburn, Clemson University
Today’s society is witnessing a tremendous increase in digital information. Myriads of applications call for the pooling and sharing of massive amounts of widely-scattered data at ever increasing scales that require a commensurate infrastructure of powerful networked distributed systems across wide and diverse areas. We will implement two existing data sharing algorithms, Cycloid and LORD, on the P2P and MANET networks, and thus identify and investigate potential issues in data sharing applications in these different heterogeneous networks. We are using GENI as the testbed for simulating the P2P and MANET network environments. Also, we will conduct a multi-system GENI experiment to demonstrate how each domain should have its own routing solutions while all the domains are federated through OpenFlow gateways.
GENI Meta Operation Center (GMOC)
Camilo Viecco, Indiana University, Global Research NOC
GMOC is now providing more operations support, including tickekting, documentation, measurements and a protected database. We have also done another emergency shutdown drill and are working to improve the process.
GENI Monitoring Slice: Enabling Network Visibility in the GENI OpenFlow Core Network (LAMP)
Ali Sydney, Raytheon BBN Technologies
One fundamental requirement of any research or production network architecture is visibility: the ability to observe the performance of a network over time. This fact becomes even more evident when abnormalities in a network become prevalent. These can include classic broadcast storms from an operator's perspective or irresponsiveness of nodes within a slice from an experimenter's point of view. To date, a plethora of network tools including iperf, nuttcap, ping, Ganglia, and Nagios are available to provide much needed insights into a network's performance. However, these tools are tailored for "barebone" network components. In GENI, users are provided "slices" which will contain some subset of virtualized, programmable computing resources. For this reason, there exist an emerging class of "visibility" tools dedicated towards monitoring, within a slice context. Among other tools, we use LAMP, adapted from perfSONAR, to manage and visualize I&M services and data. In particular, we create a reference slice which spans the GENI OpenFlow backbone to provide experimenters a somewhat "ideal" view of the network's health. Among other uses, we envision that in such cases as network connectivity irregularities within a user's slice, they can quickly refer to the reference slice as a troubleshooting guide.
GENI Research at Georgia Tech (BGPMux)
Russ Clark, Georgia Tech
Nick Feamster, Georgia Tech
Abstract General overview of GENI-related research activities at Georgia Tech.
GENI-fying Federated, Autonomous Wireless Sensor Network Infrastructures (KanseiGenie)
Anish Arora, The Ohio State University
Hongwei Zhang, Wayne State University
Rajiv Ramnath, The Ohio State University
Vipul Gupta, SUN Microsystems
Sami Ayyorgun, Los Alamos National Laboratory
Mukundan Sridharan, The Ohio State University
Wenjie Zeng, The Ohio State University
Xi Ju, Wayne State University
We present KanseiGenie, a framework for federated, autonomous wireless sensor network experimental infrastructures. KanseiGenie infrastructures provide GENI-compliant interfaces and embodies various capabilities such as programmability, virtualization, and slice-based experimentation.
Currenly KanseiGenie consortium includes four spatially distributed infrastructures: Kansei and Peoplenet at The Ohio State University, NetEye at Wayne State Univeristy, OKGerms at Oklahoma State University. As a first step towards experiment predictability and repeatability, KanseiGenie develops LENS (Language for Embedded Networked Sensing) that defines the RSpec for wireless sensor networks. LENS has been integrated with the KanseiGenie/ORCA control framework, and it has been used for federated resource management. KanseiGenie implements the embedding service for resource scheduling among federated infrastructures (e.g. intelligent control on average wireless path loss). KanseiGenie also provides users with a unified experiment control web-portal. KanseiGenie Doctor provides real-time information about the health status of the infrastructures.
We will present a live demo of KanseiGenie via a laptop, assuming there is Internet connection at the poster site.
TridentCom'10 paper on KanseiGenie: http://www.cs.wayne.edu/~hzhang/group/publications/KanseiGenie-TridentCom10.pdf
WiNTECH'11 paper on LENS: http://www.cs.wayne.edu/~hzhang/group/publications/LENS-WiNTECH11.pdf
KanseiGenie Web Portal: http://kansei.cse.ohio-state.edu/KanseiGenie/
Great Plains Environment for Network Innovation (GpENI)
James P.G. Sterbenz, The University of Kansas
Justin P. Rohrer, The University of Kansas
Egemen Çetinkaya, The University of Kansas
The Great Plains Environment for Network Innovation – GpENI is an international programmable network testbed centered on a regional optical network in the Midwest US, providing flexible infras- tructure across the entire protocol stack. The goal of GpENI is to build a collaborative research infrastructure enabling the community to conduct experiments in future Internet architecture. GpENI is funded in part by the US National Science Foundation GENI (Global Environments for Network Innovation) program and by the EU FIRE (Future Internet Re- search and Experimentation) Programme, and is affiliated with a project funded by the NSF FIND (Future Internet Design) Program.
GridStat on GENI : Simulating a Smart Power Grid Infrastructure over GENI (PlanetLab)
Divya Giri, Washington State University
Ruma Rani Paul, Washington State University
Developments in power grid measurement and monitoring technology have enabled precise and frequent measurement of the state of the power grid. Modern power grid control infrastructure are insufficient to the effective forwarding of this information to the necessary control facilities. The GridStat framework offers an efficient, low-latency data forwarding framework that can provide the necessary Quality of Service for control facilities to maintain sub-second status of monitored power grid substations. However, the current GridStat prototypes have not been tested outside local clusters. The GENI infrastructure provide the platform through which it is possible to test GridStat at scale and identify problems with the current framework.
Infinity: In-Network Storage for Mobile Devices
Yudong Gao, University of Michigan
Data accessed on the mobile devices is exploding, but current mobile applications do a poor job in conserving energy while ensuring good performance, to satisfy the rapid increase in the frequency and volume of data access. The cloud services accessed by these applications consider neither the role of mobile operator’s network nor the mobile device state, leading to poor application performance and wastage of network resources. Today’s mobile operator (MO)’s networks are no longer “dumb pipes” but are connected to data centers with large amount of resources. We argue that this disruptive change makes MO’s network increasingly resemble a cloud computing infrastructure. We propose a storage platform called Infinity that can be used by service providers to effectively exploit the mobile operator’s network, while saving energy on the mobile devices.
EAGER: Enabling Mobile Services through In-network Storage and Computation - Evaluation using the GENI Infrastructure.
Integration for GENI Experiments and Measurement Archive (OnTimeMeasure)
Prasad Calyam, OSC/OARnet, Ohio State University
Yingxiao Xu, OSC, Ohio State University
Alex Berryman, OARnet, Ohio State University
Ashiwan Sivakumar, Purdue University
Giridhar Manepalli, CNRI
Our poster describes the OnTimeMeasure measurement service integration with two GENI Experiments viz., “Resource allocation in virtual desktop clouds” led by The Ohio State University, and “Emulating cloud dynamics for performance sensitive applications” led by Purdue University. Based on these two experiment integration case studies, a general framework for “new metric” integration in OnTimeMeasure that is relevant for any GENI Experiment is presented. Lastly, the poster describes the use cases and features for archiving experiment slice measurement datasets along with meta-data collected by OnTimeMeasure into the GENI Measurement Data Archive being developed/hosted by “Digital Object Registry” led by CNRI.
Integration of LEARN with GENI Infrastructures using ORCA: VLAN Assignments and Cluster Deployment Plans - Collaborative Efforts on Measurements: IF-MAP for GENI and Collaboration with IMF
Deniz Gurkan, University of Houston
Karthik Ram Narumanchi, University of Houston
Anand Arun Daga, University of Houston
Ilia Baldine, RENCI
Rick Kagan, Infoblox
Ben Warren, Infoblox
LEARN regional optical network in Texas has been demonstrated with VLAN assignments to reach four major institutions during GEC10 (University of Houston, University of Texas at Austin, Texas A&M University, and Rice University). Planned deployment of clusters to two end points is presented (at University of Houston and Rice University). In addition, the feasibility and applicability of IF-MAP (Trusted Network Computing's Interface Metadata Access Point architecture) to the I&M services in GENI has been presented. A collaborative initiative to deploy IMF's optical physical layer monitoring software in perfSONAR to the LEARN nodes is in progress.
IMF and ORCA
Leveraging and Abstracting Measurements with PerfSONAR (LAMP)
Guilherme Fernandes, University of Delaware
Ezra Kissel, University of Delaware
Matthew Jaffee, University of Delaware
Martin Swany, University of Delaware
Jason Zurawski, Internet2
Matt Zekauskas, Internet2
Eric Boyd, Internet2
MAX GENI Aggregate Federation and Stitching
Tom Lehman, University of Southern California
Xi Yang, Information Sciences Institute, Virginia
Abella Battou, Mid-Atlantic Crossroads GigaPOP
Balu Pillai, University of Maryland
The MAX project has constructed the "Mid-Atlantic Crossroads GENI (MAX GENI) Facility" which enables the MAX Regional Network resources to be made available for GENI experiments. This includes development of a MAX Aggregate Manager which integrates the dynamic provision of network and host based resources. The host based resources include PlanetLab node virtual slices. MAX Network Stitching capabilities allow the host resources to be stitched together with Ethernet VLANs. The MAX AM is also federated with PlanetLab Princeton and ProtoGENI. In addition, a separate instance of a MAX Aggregate Manager has been deployed to "cover" the Internet2 ION Network.This combination of these capabilities now allows us to provide multi-aggregate sliver creation and stitching operations in response to Experimenters requests.
Measurement Data Archive (DigitalObjectRegistry)
Giridhar Manepalli, CNRI
Prasad Calyam, Ohio Supercomputing Center
Corporation for National Research Initiatives (CNRI) will be demonstrating the functionality of the proposed Measurement Data Archive, which is implemented using the Digital Object Architecture.
The Measurement Data Archive prototype system consists of two components: 1) User Workspace and 2) Object Archive. The User Workspace component is an entry point for users (e.g., experimenters, instrumentation researchers, etc.) to store and transfer measurement data, which could be in a variety of forms (e.g., formatted datasets, raw files, etc.). Data and metadata files managed in the user workspace can be archived for long-term storage in an Object Archive. Once data is archived, a persistent and unique identifier is created.
Network Virtualization Research In Japan
Aki Nakao, University of Tokyo / University of Utah / NICT
We have been conducting research on network virtualization infrastructure and its applications. In our attempt to pursuing building the infrastructure that may slice the network, storage and computational resources and provides network designers with systematic interfaces to reserve a slice of resources and constructing arbitrary networks on top of it, we have developed three different types of testbeds, CoreLab, VNode and WiVi. CoreLab is a software-based testbed constructed on top of COTS x86 hardware only, while VNode involves designing a new hardware based on production routers. WiVi is a recently developed WiFi virtualized testbed that may be connected with CoreLab and enable wired-wireless converged network slices. we introduce our definition of network virtualization and its applications in the context of network testbeds as wells as network architectures. We especially elaborate the detail design and implementation of our network virtualized testbeds, CoreLab, VNode, and WiVi. We also demonstrate Cache Oriented Network Architecture and OpenFlow in A Slice (OFIAS) using these infrastructures.
OpenFlow Campus Trial at Clemson: An OpenFlow Service for Seamless Enhancement of Data Transport Throughput (OFCLEM)
Aaron Rosen, Clemson University
Kuang-Ching Wang, Clemson University
Jim Pepin, Clemson University
Daniel Schmiedt, Clemson University
In a software defined network, packet forwarding methods can be changed on the fly to suit the needs of different traffic types. Not only can such a network redirect traffic's path, but it can also inject software agents in the forwarding path to provide additional services. At Clemson, we developed a solution to seamlessly enhance end-to-end data transport throughput across wide area networks. By decoupling end user and the core network's choice of transport protocols, the network provider can seamlessly enhance end users' experienced performance without requiring them to upgrade to unfamiliar new transport protocols.
Scalable Sensing Service (S3MONITOR)
Sonia Fahmy, Purdue University
Ethan Blanton, Purdue University
Sumit Kala, HP Labs
The Scalable Sensing Service (S3 Monitor) provides basic management services for users to take controlled measurements, e.g., available bandwidth or packet loss, between GENI nodes. A web interface is provided to the user for scheduling and initiating measurements, managing ongoing measurements, and retrieving measurement results.
Scalable, Extensible, and Safe Monitoring of GENI: ScalableMonitoring
Secure Content Centric Mobile Network (SECON)
Mooi Choo Chuah, Lehigh University
Xiong Xiong, Lehigh University
New wireless technologies allow mobile users to have easy access to real time data, and stay connected with friends, colleagues, & business partners. However emerging applications are usually data-centric but existing IP oriented paradigms are not flexible enough to support this. To support emerging mobile applications, we are developing a next generation mobile network that supports mobile content centric networking features, namely (a) intentional named message delivery, (b) content-centric security, (c) push-pull based data disseminations.
In our new SECON network, users can send User Interest (UI) packets to Content Resolution Server (CRS) to request for content data (CD) packets associated with a particular URI. The UIs will be forwarded by the receiving CRS to other CRSes that know who will be publishing content packets related to that URI. The UIs can also have intentional-named destinations e.g. all CRSes within a certain geographical area. In addition content publishers can send content publish announcements to CRSes before they forward content data packets to these CRSes. We have a preliminary prototype that supports UI, CPA & CD features. More features will be added in the near future.
Socially Aware Single System Image
Prof. Chunming Qiao (PI), SUNY Buffalo
Lokesh Mandvekar, SUNY Buffalo
A single system image (SSI) is the property of a system that hides the heterogeneous and distributed nature of the available resources and presents them to users and applications as a single unified computing resource. The current GENI infrastructure allows users to select and configure resources at geographically dispersed deployments and thus create their own “slice”. Our proposed experiment will allow a user to augment his resources into a single unified system by leasing/sharing resources with his friends/social contacts (along the lines of a social networking model). An SSI will thus be different from a GENI end-to-end slice in that the SSI effectively appears as a single system to the user and not as a heterogeneous end-to-end connected set of resources. An SSI cluster provides the following benefits:
- single entry point
- single user interface
- single process space
- single memory space
- single i/o space
- single file hierarchy
- single job management system
- single control point and management
The other most important benefit of using an SSI is security and data privacy. By forming an SSI with trusted resources an user can get the computing power needed to run his/her application without the fear of the application being compromised by using a third party service.
SPP Deployment and Named Data Networking Research
Patrick Crowley (PI), Washington University
Jon Turner (PI), Washington University
John DeHart, Washington University
Mart Haitjema, Washington University
Shakir James, Washington University
Jyoti Parwatikar, Washington University
Michael Wilson, Washington University
Haowei Yuan, Washington University
Washington University Internet Scale Overlay Hosting/SPP Deployment
Taiwan Integrated Research Network (iGENI)
Chu-Sing Yang, National Cheng Kung University (NCKU)
Mon-Yen Luo, National Kao Hsiung University of Applied Science (KUAS)
Te-Lung Liu, National Center for High Performance Computing (NCHC)
Robert Ricci, University of Utah
Joe Mambretti, Northwestern University
Jim Chen, Northwestern University
Fei Yeh, Northwestern University
Alan Verlo, University of Illinois, Chicago
Maxine Brown, University of Illinois, Chicago
Tom DeFanti, University of California, San Diego
One of the international partner projects for iGENI in Taiwan, the National Science Council/Taiwan funded the National Telecommunication Project: Study and Deployment of Network Virtualization Architecture(NCKU,KUAS and other universities) has initiated the development and deployment of a new network virtualization architecture on a national research/education backbone: TWAREN (NCHC). In addition to this collaboration project, iGENI also worked with ProtoGENI team to implement a direct connection between ProtoGENI and network research infrastructure in Taiwan, which is enabling an enhanced partnership between GENI community and the Taiwan network research communities.
The Performance Evaluation of Bandwidth Allocation Algorithms in Multi-domain Networks
Jiten Patel, Rochester Institute of Technology
Kaiqi Xiong, Rochester Institute of Technology
This project is concerned with dynamic bandwidth allocation in multi-domain networks. We have been designing and developing the algorithms to minimize the total cost of network bandwidth when satisfying the requirements of Quality of Service (QoS) predefined in the Service Level Agreement (SLA). An SLA is a contract negotiated between a network service provider and customers. The goal of this research is to evaluate the performance of these bandwidth allocation algorithms by a use of the GENI infrastructure. We have conducted experiments for the validation of percentile delay calculations as well as the evaluation of Additive Increase/Multiplicative Decrease (AIMD)-based bandwidth allocation algorithms. Furthermore, by using the research experiments of this project, we have designed GENI educational experiments that have been used in networking courses at RIT.
Alvin AuYoung, HP
Andy Bavier, PlanetWorks?
Jessica Blaine, HP
Jim Chen, Northwestern University
Yvonne Coady, University of Victoria
Paul Muller, University of Kaiserslautern
Joe Mambretti, Northwestern University
Chris Matthews, University of Victoria
Rick McGeer, HP
Chris Pearson, University of Victoria
Alex Snoeren, UC San Diego
Fei Yeh, Northwestern University
Marco Yuen, PlanetWorks?
Transcontinental federation of cloud systems.
Trema; An Open Source OpenFlow Controller Platform
HIDEyuki Shimonishi, System Platforms Research Laboratories, NEC Corporation
Yasunobu Chiba, System Platforms Research Laboratories, NEC Corporation
Yasuhito Takamiya, System Platforms Research Laboratories, NEC Corporation
Kazushi Sugyo, System Platforms Research Laboratories, NEC Corporation
- Trema is a free OpenFlow controller platform (GPL v2)
- Assists anyone who wants to develop his/her own OpenFlow controller
- Not targeted for any specific OpenFlow controller implementation
- Trema allows to implement OpenFlow controllers in C and Ruby
- Trema provides:
- Various basic libraries on which you can build your own OpenFlow controller
- Integrated network emulator and developing environment
- Mailing list: firstname.lastname@example.org / twitter: @trema_news
- Mailing list: email@example.com / twitter: @trema_news
TUNIE: A Flexible and Programmable Virtualized Network Innovation Environment in China
Yong Li, Electronic Engineering, Tsinghua University
Network community needs a flexible platform for network experiment of new architectures, algorithms and protocols in the research of network innovation. However, building such a platform faces lots of challenges due to its complicate requirements. In this poster, we present TUNIE, a network testbed for rapid concurrent experiment of network innovation on virtualized programmable infrastructure in China. ExpoNet provides end-to-end slice including wired and wireless components, which integrates both software- and hardware-based router virtualization technologies to provide a flexible approach to configure and customize both the control plane and data plane while satisfying various experiment requirements. In the wireless part, we have a sensor testbed including 100 wireless sensor nodes, and a WiFi testbed. In the wired part, we have setup one OpenFlow network, and two virtualization testbed based on multi-core servers and FPGA data plane. In our current platform implementation, we have four sites, two sites in Tsinghua University, one another university of BUPT, and one in China Union, one of the largest Service Providers in China. We have setup a Federation plan to extend our platform with other Universities and companies like HUST, Huawei, etc.
Using OMF/OML for WiMAX Experiments (WIMXBBN)
Abhimanyu Gosain, Raytheon BBN Technologies
Harry Mussman, Raytheon BBN Technologies
Hamed Soroush, Raytheon BBN Technologies
This poster presents work on integration of the WiMAX base station and clients with the Orbit Management Framework (OMF) and the Orbit Measurement Library (OML). We present results from network experiments instrumented and conducted using tools provided by OMF.
Wide-area programmable network: Operation And Management of a nation wide Openflow-Based Network in Japan
Shuji Ishii, NICT
The focus of this study is on developing a management technology of global OpenFlow-based networks integrated into the existing VLAN-based networks. We deployed a nation-wide OpenFlow-based network on the NICT JGN-X testbed in Japan. And we demonstrated a trial high-quality realtime video streaming application to verify the performance and reliability of large-scale OpenFlow-based networks. We also describe the details of technical issues confronted in the trial operation.