National Laboratory for Applied Network Research

Program Plan

October 1, 1996 - September 30, 1997

Cooperative Agreement No. NCR-9796
between the
National Science Foundation and the University of California, San Diego

I. Contents

II. Overview of NLANR

The NLANR Cooperative Agreement includes tasking for the following areas:

  1. Technical and engineering support and overall coordination of the vBNS connections at the five supercomputing centers and selected Research and Education sites.

  2. Testing and measurement of the vBNS performance characteristics which have been agreed upon jointly by the NSFNet Program Official and the vBNS Awardee [MCI].

  3. Coordination and oversight of the use of the vBNS as a shared facility among the supercomputing centers, selected Research and Education sites and the Awardee [MCI]. This use as a shared facility should not conflict with its primary intended use as a research platform.

  4. Coordination and scheduling of utilization of the vBNS by researchers identified and referred to the SCCs. [Note that this tasking has expanded to encompass NSF Connections Program awardees and other meritorious sites -- namely those vAIs and vPIs approved by NSF/DNCRI.]

  5. Participation in the Research Allocation Committee (VTCC) for the vBNS. [While the vBNS Technical Coordinating Committee (VTCC) is active currently, the originally conceived RAC is not.]

  6. Coordination of activities at the Supercomputing Centers and selected Research and Education sites related to the enforcement of the vBNS Acceptable Use Policy and dissemination of related information. [This program plan expands this tasking to include direct support of the SCCs and select R&E sites.]

During the May 1995 through September 1996 period of this cooperative award, NLANR activities were slower to gear up than originally anticipated. This fact was due in part to technical/policy delays associated with the use of the vBNS (particularly with respect to the appropriate Acceptible Use Policy for the network) and contractual delays associated with the transfer of the Cooperative Agreement from General Atomics to the University of California, San Diego (UCSD).

During 1995/1996, NLANR activities focused in two distinct but related goals:

  1. Helping NSF's Division on Networking, Communications, and Research on Infrastructure (NCRI) to provide a viable network environment to the research and education (R&E) community, in order that scientists can take maximum advantage of community computational resources -- many of which are also provided by NSF (i.e., the supercomputers (vector and mpp), visualization/graphics engines, workstation clusters, data archives, mass storage, and AFS services of the NSF-sponsored supercomputer centers).

  2. Using the vBNS and the SCCs as a testbed for exploring critical technologies and analysis methodologies which have potential to upgrade the Internet community's understanding of fundamental traffic behavior, e.g., metrics measurement and analysis tools and traffic visualization techniques. The knowledge developed through this type of research may serve as the foundation for classes and qualities of service, accounting, and related economic models, as well as enhance the ability to engineer next generation networks.

NLANR has supported these goals through at least four separate capacities:

  1. Helping R&E users effectively run their applications on the vBNS;

  2. Providing technical and operational support for NSF connections equipment and configuration at each site;

  3. Pursuing and coordinating forward-looking research and applying these efforts to NSF's research infrastructure; and

  4. Providing technical feedback relevant to NSF connections policy issues.

A significant portion of NLANR's resources at Cornell Theory Center (CTC), the National Center for Atmospheric Research (NCAR), and the National Center for Supercomputing Applications (NCSA), and Pittsburgh Supercomputing Center (PSC) have, over the past year, been devoted to technical support (item 2.). For the remainder of the NLANR cooperative agreement, NLANR participating sites will increase their activities towards supporting the larger NSF high performance connections community (item 1) and expanding related research (item 3). Site personnel will continue to provide ongoing support related to technical and policy issues relating to vBNS connections (item 4).

During FY1997, NLANR will leverage the experience of the initial vBNS sites over the last two years, providing a more proactive approach to supporting the vBNS users, particularly the 50+ institutions expecting to be connected to the vBNS during FY1997 as part of the NSF Connections Program. NLANR will also expand its highly acclaimed research efforts relating to measurement, caching, and visualization.

III. FY1997 Topic Areas

NLANR's efforts in FY1997 are divided into three topic areas:

  1. User Services / Outreach - user support, tracking and training. Currently, only 1.9 FTEs are appropriated to this Topic Area by the five supercomputing centers (SCCs). This level of effort is consistent with assisting personnel at each of the supercomputing centers with implementing their applications on the vBNS. The types of applications- and site- specific support provided under this Topic Area are reflected in past NLANR progress reports.

    In order to broaden this support to encompass the anticipated 50+ new NSF Connection Program sites connecting to the vBNS this year, the level of effort under this Topic Area will need to be increased. This will be addressed in large measure through a separate proposal anticipated from NCSA to NSF in Spring 1997. A description of illustrative User Services which might be supplied under this a higher level of support is provided in Attachment 1.

    FTE (full time equivalent) resources to be applied under the current NLANR funding levels include:

    TOTAL: 2.45 FTEs

    Additional resources required: Additional resources may be requested by NCSA through a future proposal, separate and distinct from the NLANR cooperative agreement.

  2. Engineering - technical/operations support coordination. This area will be led by Jamshid Madahvi, PSC. Engineering support will encompass coordination of technical discussions, i.e., planning, general meetings of new sites, perhaps co-located with related meetings, such as those of NANOG, IETF, and Internet-2. NLANR's Engineering Working Group will also monitor the performance of applications running currently over the vBNS and new connections research infrastructure concentrated mainly on the large number of un-instrumented applications that have no way to determine what level of network bandwidth they are consuming. This working group will also be responsible for forging technical exchanges with CA*net2/NTN, DREN, ESNET, and other high performance network infrastructure groups during FY97.

    Specific activities which will be addressed by each of the participating NLANR sites include the following:

    1. assisting MCI Engineering with the installation, testing, and operation of the vBNS performance monitors;

    2. assisting MCI Engineering with the installation, testing, and operation of the OC-12 network circuit upgrades;

    3. assisting MCI Engineering with the installation, testing, and operation of Fore Systems ASX-1000 ATM switch hardware and software at each SCC; and

    4. providing some assistance to Duane Wessels with the implementation of the NLANR Web caching hierarchy.

      5. FTE resources include:

        Cornell - .2
        NCAR - .45
        NCSA - .25
        PSC - 2
        UCSD - .1

      TOTAL: 3.0 FTE

    Additional resources needed: None anticipated.

  3. Research - inc. measurement tools, caching, Mbone, and IP v.6. k claffy has held the position of NLANR's Research Coordinator since its inception. She will continue to lead this topic area. Specific activities planned for FY1997 include: enhancement of traffic visualization tools, development/deployment of measurement and analysis tools; continued development of the international caching hierarchy; IP v.6 experiments (in addition to continuing Mbone efforts); and promotion of collaborative network environments. Details on these topics are provided in the UCSD section below.

    Resources include:

    TOTAL: 6.1 FTE

    [Note that the resources included under the UCSD category includes Tracie Monk (75%) for outreach specific to the NLANR research program (e.g., measurement and caching), administrative support related to meetings and general NLANR activities, and internship support from individuals such as Tony Sterret who is porting OC3mon to FreeBSD and Carl Maeda who is supporting visualization efforts. It also includes .5 FTE for a webmaster to perform work previously done by the research coordinator.]

    Additional resources needed: No additional resources are anticipated beyond those specified in the January-September 1997 budget submission.

In addition to these three Topic Areas, resources will also be applied to overall management of NLANR and to providing NSF with Technical Feedback on Policy, as needed. This support will primarily be provided by the Principal Investigators associated with each of the SCC sites and is presumed to be included in their .1 FTE allocation for management.

In addition, all sites will continue to participate on vBNS Technical Coordination Committee conferences and attend quarterly NLANR/vBNS meetings.

IV. Site-Specific Plans

1. MCI vBNS Plans for FY1997

The primary focus of NLANR's User Services and Engineering Support activities will be associated with expansion of the vBNS and its connections. A summary of MCI's plans for the vBNS during FY1997 is included below. Additional details on the vBNS cooperative agreement between NSF and MCI are available at http://www.vbns.net.

As previously mention, RSVP will continue to be tested in the vBNS testbed, with RSVP-services deployed in the vBNS as the router implementations provide a useful service.

Other plans for FY97 include improved multicast services with functionality extended to the provisioning of point-to-multipoint VCs with a possible MARs/MCS server deployed for the vBNS.

Data visualization work will leverage NLANR's mbone, web caching, and AS topology visualization, as well as their end-to-end path performance visualization tools, to produce similar capabilities for the vBNS. We will also work on improved visualization of the OC3mon flows data and the SNMP-based data.

2. Cornell Theory Center (CTC)

User Services: CTC will continue its work with sites to facilitate their efforts to seek/implement connections to the vBNS through NSF Connections awards. [Note Cornell assisted five such sites in 1996.]

CTC will continue its work with CLEO -- a collaboration of over 200 high energy physicists from 24 universities studying the production and decay of heavy quarks and leptons produced in the Cornell Electron Storage Ring (CESR). Specifically, CTC will work with Michael Ogg and other CLEO personnel to identify routes to CLEO sites which could be improved by routing through the vBNS and ESNet. CLEO has extensive records of test results with reachability and round trip times between Cornell and other sites for future comparisons (see: http://www.nile.utexas.edu/Nile/components/networks/CLEO-sites/Cornell/) . CTC would then seek approval from NCRI and, while coordinating with ESNet, manually configure Cornell routers to accept these routes. CTC will also work with ESNet personnel and MCI to assist in automating the routing by utilizing the Communities Attribute in BGP.

Engineering Services: CTC will continue testing of Cisco's beta RSVP code and FORE's IP Multicast code and beta Classical IP over ATM code. CTC plans to develop better mechanisms for scheduling bandwidth and to integrate network bandwidth reservation with distributed computing. It will also conduct experiments and interoperability testing of desktop videoconferencing applications, specifically: vic and vic-compatible implementations under IRIX, AIX, Mac OS, and Windows. In preparation for these and other projects, CTC is re-architecting internal networks and connections to the SP2 using routed and switched ethernets and IP over ATM links, integrating a IBM 8260 ATM switch which attaches production AFS servers with a FORE 7000 PowerHub, FORE ASX-1000 ATM switch, ethernet switches, and cells-in-frames (CIF) attachment devices. Cornell will experiment with and evaluate several alternatives for virtual routers between Classical IP and Emulated LANs- including a workstation-based solution, FORE's beta code for the PowerHub, and an IBM 8210 Multiprotocol Switched Services server.

Research: No specific activities are planned.

3. National Center for Atmospheric Research (NCAR)

During FY1997, NCAR will provide engineering and technical support and coordination for the vBNS connections at NCAR and will provide vBNS connection related support to the University of Colorado at Boulder.

User Services/ Outreach: Anticipated user service support will include: assisting the UCAR UNIDATA Program with Internet Data Distribution project over the vBNS and assisting the NCAR Mesoscale and Microscale Meteorology Division with a distributed mesoscale modeling application over the vBNS.

Engineering: Engineering and technical support efforts will include support as described in the Engineering Services section (above). It will also include engineering support to the University of Colorado, Boulder and its (gigaPoP-like) connection at NCAR.

Research: No specific activities are planned.

4. National Center for Supercomputing Applications (NCSA):

User Services/Applications Support/Outreach: NCSA's .5 FTE will be focused on developing and holding a vBNS Applications Workshop is planned for Spring 1997 which will focus on high performance networking technologies and issues such as ATM, RSVP, MPI and achieving high performance over high-speed, high-latency networks. The workshop outline is at http://www/People/rbutler/workshop.html. Other user service activities will include recruiting new applications to be run over the vBNS and to helping to support and demonstrate select applications at Supercomputing '97. NCSA's support for vBNS applications includes technical assistance relating to benchmarking, optimizations, problem resolution, and special configurations.

Engineering: This is the minimum it will take to support all the technical support issues. This includes, problem resolution, routing, monitoring, etc. Additionally we developed a route server tool set to look at what routes the vBNS is handing customers and we plan to expand its functionality in 1997. The WWW page for this is in progress at http://rs.ncsa.uiuc.edu.

Research: NCSA is participating in research efforts to develop/deploy the OC3mon and OC12mon flow monitors. Specifically, NCSA is responsible for porting OC3mon to LINUX with local monitoring extensions and for preparing a partial rewrite of the flows software, including related documentation.

NCSA intends to submit a supplemental proposal to NSF to cover additional resources relating to User Services. This proposal will request an additional FTEs to provide support related to User Services tasks, see the Attachment.

5. Pittsburgh Supercomputing Center (PSC)

For the remainder of the NLANR cooperative agreement, PSC will gear its activities towards supporting the larger community of NSF high performance connections awardees. These efforts will include both engineering efforts, working with new sites on the vBNS and high performance networking technical issues, and user services/outreach efforts, focused on end users at high performance connections sites.

User Services/Outreach: The planned user services and outreach activities include: establishing direct contacts with users at new high performance connections sites; working with these users on the development of high performance networking applications; and providing documentation for users on techniques for optimizing network performance. In addition, PSC will identify high demand users who are not currently at High Performance Connections sites and provide them with the necessary information to request high performance connections should it be justified.

Engineering: The technical and engineering effort at PSC will focus on working closely with MCI and other sites on testing and deploying new networking technologies and capabilities in the vBNS. Over the coming months, we see significant effort devoted to routing for VAIs and VPIs; monitoring for both overall performance of vBNS connections and individual flow performance information; testing different implementations of QoS capability in the vBNS; upgrading and maintaining vBNS MBone infrastructure; providing technical enhancements required to enable the use of multimedia applications on the vBNS including audio and video conferencing; and connecting to and testing vBNS IPv6 infrastructure.

Research: No specific activities are planned using NLANR funds. However, PSC's SACK and NIMI efforts are intimately related to, and will be coordinated with, NLANR research activities at UCSD.

6. University of California, San Diego (UCSD):

User Services: No specific activities are planned.

Engineering: Engineering and technical support efforts will include support as described in the Engineering Services section (above).

Research: UCSD will provide support in the following areas:

New efforts will include IP v.6 experiments in cooperation with MCI and participants of the Collaborative Advanced Interagency Research Network (CAIRN).

V. Progress Reports

Information on NLANR's past activities is available in the following progress reports:

VI. Attachment:

SUPPLEMENTAL USER SERVICES SUPPORT

The items below are a brief summary of the types of support which NCSA envisions might be provided to vBNS users under a proposed cooperative agreement between NCSA and NSF. Note that these proposed User Services are not covered under the existing NLANR cooperative agreement.

Last updated 28 February 1997

Questions should be directed to info@nlanr.net