Changes between Version 3 and Version 4 of GeniDesktop_year2_report


Ignore:
Timestamp:
08/11/15 17:05:25 (9 years ago)
Author:
griff@netlab.uky.edu
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • GeniDesktop_year2_report

    v3 v4  
     1[[PageOutline]]
     2
     3= GENI Desktop Project Status Report =
     4
     5Period: Solicitation 4 Year 2
     6
     7== I. Major accomplishments ==
     8
     9The following highlights our accomplishments during the last year.
     10
     11=== A. Milestones achieved ===
     12
     13 * Modified the GENI Desktop to support "Speaks-for" authentication being developed/supported by the control frameworks and other GENI tools/services. 
     14
     15 * Incorporated user-driven feedback into the GENI Desktop to support new user-requested services and features. 
     16
     17 * Developed new training materials that incorporate the changes made to the GENI Desktop.
     18
     19 * Integrated and leveraged existing tools and services (e.g., Jacks) into the GENI Desktop for managing topologies, experiments, and results.
     20 
     21 * Collected user feedback regarding usability of the GENI Desktop and made major changes to improve its ease-of-use and aesthetics.
     22
     23 * Adapted the GENI Desktop archiving service for storing and retrieving experiment results and artifacts from iRoDs to support a new archival service with enhanced features.
     24
     25 * Enhanced the set of scriptable resource management, instrumentation, and monitoring available to experimenters and other tools.
     26 
     27 * Enabled integration of the GENI Desktop with other experimenter tools.
     28
     29 * Created documentation and tutorial materials to reflect this latest version of the GENI Desktop.
     30
     31=== B. Deliverables made ===
     32
     33 * We enhanced the GENI Desktop to use "Speaks-for" credentials for accessing resources from other GENI components on behalf of users.
     34
     35 * We implemented an initial version of the slice verification and configuration testing service.
     36
     37 * We developed code for super slice support in the GENI Desktop.
     38
     39 * We developed a completely new user interface for the GENI Desktop which we call "GENI Desktop Lite" that greatly improves the look-and-feel and ease-of-use of the GENI Desktop.
     40
     41 * We integrated the Jacks tool into the GENI Desktop, enabling users to create topologies and instantiate experiments (i.e., slices) using the Jacks tool.  We also integrated the Adopt-A-GENI (AAG) tool into the GENI desktop.
     42
     43 * We developed and integrated a new archival service into the GENI Desktop that leverages VMs to hold, and later display, the archived experiment state in the same context as it was initially collected and viewed.
     44
     45 * We designed and implemented a GENI Desktop Command Line Interface (gdcli) that enables users to write scripts that control, manage, and measure the performance of their slices through the GENI Desktop.
     46
     47 * We demonstrated how the new gdcli can be used by other experimenter tools to integrate with the GENI Desktop.
     48
     49 * We developed online documentation for the new gdcli interface and gave a tutorial entitled "Monitoring and Controlling Experiments with GENI Desktop Scripts and Modules" at the GEC 23 conference.
     50
     51== II. Description of work performed during the last year ==
     52
     53The following provides a description of the progress made during the last year.
     54
     55=== A. Activities and findings ===
     56
     57Our activities this past year have resulted in enhanced functionality,
     58improved ease-of-use, better authentication/security, a redesigned user
     59interface, and the ability to control the GENI Desktop programmatically
     60throught scripts.  In particular, we:
     61
     62 * Incorporated support for "Speaks-for" into the GENI Desktop
     63
     64 * Designed and implemented a slice verification service
     65
     66 * Designed and implemented a super slice abstraction
     67
     68 * Developed a new archival services based on Xen VMs to restore entire contexts
     69
     70 * Redesigned the look-and-feel of the GENI Desktop user interface to make it much easier to use
     71
     72 * Integrated Jacks and Adopt-A-GENI (AAG) tools into the GENI Desktop
     73
     74 * Designed and implemented a GENI Desktop Command Line Interface (gdcli) to enable script-based access to GENI Desktop functionality.
     75
     76The following briefly describes our activities this past year.  We begin with our efforts to
     77improve the authentication/authorization used to access the GENI Desktop, and
     78then move on to describe new functionality added to the GENI Desktop,
     79followed by a description of our efforts to enhance the look-and-feel of the
     80GENI Desktop, and
     81lastly our efforts to allow scripting of the GENI Desktop.
     82
     83
     84==== Authorization: Supporting "Speaks-for" ====
     85
     86The "Speaks-for" credential allows trusted tools to act for, instead of
     87acting as, an experimenter to perform certain actions, such as requesting
     88resources from aggregates, accessing allocated resources, and installing
     89software on experimental nodes. We enhanced the GENI Desktop to use "Speaks-for"
     90credentials for accessing resources on behalf of users. Users no longer
     91need to provide the private key to the GENI Desktop. We implemented an interface
     92for the user to authorize the GENI Desktop to speak for her/him.  A GENI
     93Desktop-specific certificate is signed using the private key of the user.
     94Because the whole process happens within the browser on the client side,
     95the private key never leaves the user's machine. The "Speaks-for" credential
     96allows the GENI Desktop to talk to aggregates and perform all necessary
     97actions on behalf of the user.
     98
     99==== New Functionality (1): Jacks, Archival, Verification and Super Slice Services ====
     100
     101Another goal this past year has been to begin integrating support for other tools
     102into the GENI Desktop.  To demonstrate the ability to support other tools, we
     103began by integrating the Jacks tool into the GENI Desktop.  Users can now add
     104resources to their slices by selecting Jacks in the GENI Desktop which will
     105direct them to a GENI Desktop page that embeds the Jacks tool and allows them
     106to allocate the resources (i.e., which uses the OMNI tool).  RSPECs
     107created by Jacks can be saved by the GENI Desktop for future use.
     108
     109In addition, we integrated the Adopt-A-GENI (AAG) flow specification module into
     110the GENI Desktop, allowing users to visually define OpenFlow paths across the
     111topology that are then sent to the AAG module to be instantiated in the
     112OpenFlow controller.  Although the AAG functionality is logically a distinct
     113service/tool, the messaging system between windows in the GENI Desktop made
     114it possible to incorporate this new tool with relatively little effort.  In
     115addition, we were able to add a new AAG Controller node type to the Jacks
     116wrapper, thereby integrating the AAG controller into the Jacks tool as well.
     117
     118The existing archival service in the GENI Desktop leveraged the iRoDs storage
     119service to store and later retrieve measurement data collected by the GENI
     120Desktop.  A key limitation of this service was the inability to easily (and
     121quickly) access, view, and make sense of archived measurement data.  To
     122address this need, we developed a new archival service that not only archives
     123the measurement data, but also archives the software and context used to
     124display the data.  Because the data and the environment needed to view the
     125data are archived, users can quickly access an archive and view the saved
     126data using the same tools available at the time the data was collected.
     127
     128To support this new archival service, we implemented an archival server that
     129not only captures the measurement data stored on the global node (where
     130measurement data is collected), but it also captures the state of the drupal
     131system used to display the data, including all web server (apache) and
     132database (mysql) files.  GENI Desktop users can request that an archive be
     133made, which is then sent to the archive server.  When a user visits the
     134archive web page on the archive server, they can select from any of the
     135archived snapshots.  The archive server will dynamically launch a Xen VM,
     136setup the apache, mysql, and Drupal state needed to view the measurement
     137data, install the archived measurement data, create login credentials for the
     138user, and share the credentials with the GD so the user is automatically logged
     139into the archive VM.  The result is that the user is presented with the same
     140look-and-feel as if they had gone to the global node at the time the snapshot was taken.
     141
     142We implemented the slice verification and configuration testing service
     143as a module in GENI Desktop by taking advantage of the module builder function
     144of the GENI Desktop. Based on the manifest of an experiment, the verification
     145service analyzes the topology and performs tests about the interfaces of all
     146nodes in the experiment. The initial version we implemented checks whether
     147each interface is up and whether it is reachable from a ping test.
     148The results are presented in a table showing the status of all the interfaces of
     149all the nodes in the experiment.
     150Later versions of the verification service included
     151additional checks (particularly automated bandwidth checks) and also made it
     152possible for users to write their own verification scripts to test for things
     153of importance to their experiment. 
     154
     155Building a large experiment is a difficult task in GENI, partly because
     156it is more likely to fail if we create an experiment with a lot of nodes.
     157At the same time, we may have multiple related experiments and want to
     158combine these relatively small experiments together to form a large experiment.
     159We developed a new "super slice" service in the GENI Desktop to support this functionality.
     160Users can use the GENI Desktop to create a super slice by combining multiple
     161existing slices together. The GENI Desktop provides a GUI for users to display
     162multiple slices at the same time and pick any pair of nodes from different
     163slices to establish a link between them. The Super Slice service in the GENI
     164Desktop currently can then automatically set up GRE tunnels between these selected pairs of nodes
     165from different slices.
     166
     167==== A New User Interface: GENI Desktop Lite ====
     168
     169Over the years the number of features and capabilities offered by the GENI
     170Desktop has continued to expand.  Indeed, a key goal of the GENI Desktop was
     171to provide users with a context for managing all aspects of their experiment
     172from setup and deployment to monitoring and archiving of measurements and
     173results.  The downside to this expanded functionality is increased complexity
     174using the tool.
     175At the same time, the number of experimenters who are using the GENI Desktop to
     176create, manage, monitor, and control their slices has been grown rapidly,
     177due, in part, to users being exposed to the GENI Desktop as part
     178of GENI tutorials, summer camps, demontrations and online documents and
     179videos.  Feedback from this user group indicated that the extensive
     180functionality available in GENI Desktop made it difficult for new users to
     181navigate and use.
     182
     183To address this need for a tool that could be easily learned and used by new
     184users, we completely redesigned the look-and-feel of the GENI Desktop to
     185reduce complexity and make it simple to create, run, and monitor
     186experiments.  Our new "GENI Desktop (GD) Lite" interface is now the default
     187intereface that users see when they log into the GENI Desktop.
     188Users can still access the (original) advanced user interface if needed, but in most
     189cases find that the GD Lite interface is sufficient.
     190The Lite intereface is designed to take users through the lifecycle of an
     191experiment.  The Lite intereface starts by helping users create a slice,
     192assigning resources to the slice, and then giving them access to a simplified
     193version of the GENI Desktop topology view where they can log in to nodes, run
     194their experiment, monitor basic traffic types, and archive results.  Initial
     195feedback on the new intereface has been extremely positive.
     196
     197In addition to a major rewrite of the web code for the user interface, one of the key
     198challenges that we had to address was automating the global node setup, initialization,
     199and instrumentation.  While these "backend" operations were clearly visible
     200in the old user interface, they had to be hidden in the new interface.  This
     201meant that the GENI Desktop had to be able to add global nodes into the slice
     202(one for each aggregate) on the user's behalf.  This required working with
     203the aggregates to support the GENI AM API calls needed to add resources to an
     204existing slice.  In addition, the GENI Desktop needed to be able to initialize and then
     205instrumentize the slice in the background (i.e., while allowing the user to
     206view and use the slice in the GENI Desktop).  This required changes to the
     207GENI Desktop to monitor the background initialization/instrumentation
     208process and incrementally enable functionality as it became available.  For
     209example, while resources are being allocated the GENI Desktop can only display
     210the topology and the status of the node initialization.  As soon as the
     211initilization completes, the file upload, ssh, and run command functionality become available in the user
     212interface.  Later when the instrumentation completes, functionality such as
     213displaying basic traffic graphs or archiving measurement data become
     214available in the user interface.  In short, users are now taken directly to the
     215GENI Desktop topology view, bypassing several setup steps required by the old
     216user interface.  Commonly needed functionality is then automatically added as
     217it becomes available.   As part of the new Lite interface, we also simplified
     218the design of the web page(s) used to select a predefined RSPEC.
     219
     220==== Programming the Desktop:  The GENI Desktop CLI (gdcli) ====
     221
     222The GENI Desktop greatly simplifies the task of instrumenting and monitoring
     223a users' experiment (slice).   However, users could only access the GENI
     224Desktop via a web interface.   In other words, there was not programmatic way
     225for experimenters or tool developers to leverage the GENI Desktop
     226functionality.
     227
     228To address this need we designed a new interface to the GENI Desktop that
     229could be used to programatically upload files, run commands, download
     230measurement graphs, etc --- functions previously only possible via the GENI
     231Desktop web interface.  In particular, we developed an application that runs
     232on Linux (or other Unix-based systems), Mac, and Window called the gdcli
     233program that can be used to interact with the GENI Desktop.  The gdcli
     234program can be used to:
     235
     236 * Upload files to a select set of nodes
     237 * Run a command on a select set of nodes
     238 * Download a traffic measurement graph (as PNG or CSV) from a select set of nodes
     239 * Download a normal file from a select set of nodes
     240 * Get a list of slices
     241 * Check the status of a slice
     242 * Get the topology of a slice
     243 * Validate the setup of a slice
     244 * List the nodes in a slice
     245 * List the links in a slice
     246
     247The gdcli program can be called from any scripting language (e.g.,
     248python, perl, sh (bash), .BAT files, etc).  As a result, users are able to
     249write programs in their favorite scripting language that make calls to the
     250GENI Desktop to upload/download files, download measurement graphs, run
     251commands, etc.
     252
     253There were several challenges that we had to address
     254while implementing the gdcli scripting interface.
     255First, we needed a way to make calls to the GENI Desktop server (e.g., to
     256download a traffic graph, or run a command).  To solve this problem we
     257enhanced the GENI Desktop server to support HTTP posts that included
     258parameters to the request specifying, for example, the list of traffic graphs to be
     259downloaded (i.e., the nodes/links names and the types of graphs desired).
     260We implemented a python backend server specifically designed to process the
     261request, perform the action, and return the results.
     262The python backend shares access
     263with the previous GENI Desktop PHP code to the databases and files used by the
     264GENI Desktop, thereby ensuring that the results returned by the gdcli are the
     265same information as would be seen in the GENI Desktop web interface.
     266
     267A second challenge was securing the access to, and communication with, the
     268new python backend server.  To ensure communication is secure, all communication
     269occurs over a secure connection using https.  The problem of authorization
     270requires not only that the user authenticate themselves to the server, but
     271that the server obtain a "speaks-for" certificate to act on behalf of the
     272user.   Because the existing speaks-for generation tools are designed for
     273interactive web use, not scripting, we decided to require that users first
     274authorize a speaks-for using the existing GENI Desktop web interface which
     275can then be stored and used by the GENI Desktop (and our new backend server)
     276until the speaks-for expires.  However, this does not solve the authorization
     277problem. To ensure the users has the right to issue commands to our python
     278backend server, the web interface of the GENI Desktop also creates a secret
     279key (say at the same time the user authorizes the speaks for) that the user
     280must store on their local machine.  The secret key is used when communicating
     281with the python backend server to prove that the user has the right to invoke
     282the requested operations on the GENI Desktop.  In that sense, users can think
     283of the gdcli secret key like an ssh key that must be present on their local
     284machine in order to access the service.
     285
     286A third issue involved handling the results/output of a gdcli request.  The
     287gdcli tool provide two mechanisms for handling the output from a request.
     288The first, and most simple mechanism, concatenates all the output files/graphs and prints
     289them to standard output, allowing users to redirect output to other programs
     290or tools. The second way gdcli handles output is to deposit each graph,
     291downloaded file, or output from a run command into a different file on the
     292local machine.  Files are automatically assigned names that describe their
     293content (based on the slice, the aggregate, the node or link, and the type
     294of graph).  Because the naming convention is known to experimenters, they can
     295easily write scripts that know what filenames to look for, and then feed
     296those files to the appropriate program for processing (e.g., copying traffic
     297graphs into a web directory to create a user-defined traffic mashup view).
     298
     299=== B. Project participants ===
     300
     301The following individuals are involved with the project in one way or another:
     302 * Jim Griffioen - Project PI
     303 * Zongming Fei - Project Co-PI
     304 * Hussamuddin Nasir - Technician/Programmer
     305 * Charles Carpenter - Technician/Programmer
     306 * Xiongqi Wu - Ph.D. Student
     307 * Jeremy Reed - Ph.D. Student
     308
     309=== C. Publications (individual and organizational) ===
     310
     311=== D. Outreach activities ===
     312
     313 * We gave a presentation about the GENI Desktop and its features during the Introduction to GENI Instrumentation & Measurement Tools portion of the Getting Started with GENI tutorial at GEC 22 and GEC 23.
     314 
     315 * We gave a demo of the latest GENI Desktop features during the demo session at GEC 21, GEC 22, and GEC 23.
     316 
     317 * We gave a tutorial entitled "Monitoring and Controlling Experiments with GENI Desktop Scripts and Modules" at the GEC 23 conference.
     318 
     319 * We developed and posted online-documentation and online-tutorials that describe the new features of the GENI Desktop for users.
     320 
     321=== E. Collaborations ===
     322
     323 * Most of our collaborations have been between the GPO Portal team and the aggregate teams at Utah and RENCI.
     324
     325=== F. Other Contributions ===
     326