National Laboratory for Applied Network Research

Final Report

October, 1998

Cooperative Agreement No. NCR-9796124
between the
National Science Foundation and the University of California, San Diego

NLANR Final Report:

Introduction

The National Laboratory for Applied Network Research was a collaborative effort among the five NSF-sponsored supercomputing centers (Cornell Theory Center (CTC), National Center for Atmospheric Research (NCAR), National Center for Supercomputing Applications (NCSA), Pittsburgh Supercomputing Center (PSC) and the University of California, San Diego / San Diego Supercomputer Center (UCSD/SDSC). It was supported by the Division of Networking and Communications Research Infrastructure of the National Science Foundation. A primary objective of NLANR was to support researchers on the NSF/MCI very high speed Backbone Network Services (vBNS), a national network research vehicle that connected the five SCCs at high bandwidth.

The specific work performed under the NLANR agreement included technical and engineering support and overall coordination of the vBNS connections at the five supercomputing centers, support of NSF High Performance Connections sites, as well as testing and measurement of the vBNS and related Internet performance characteristics.

NLANR's primary goal was to provide technical, engineering and traffic analysis support of NSF High Performance Connections sites and the broad vBNS user community.

Throughout the NLANR project, all sites provided ongoing engineering and general support related to the vBNS operations and applications and participated in the NLANR caching project Global Caching Hierarchy. Past NLANR Quarterly Reports can be found at: NLANR Quarterly Reports

Goals

The National Laboratory for Applied Network Research (NLANR) was a collaboration among NSF-supported supercomputer sites. NLANR was created in 1995 to provide technical and engineering support and overall coordination of the vBNS connections at the five NSF-supported supercomputer centers. The vBNS evolved to become a "leading edge but stable" platform to enable the development and use of high performance applications by the broader academic research community. NLANR's focus expanded beyond the initial engineering effort into providing user support for network projects and conducting network research, particularly in measurement and operations analysis of network traffic.

NLANR's initial engineering focus expanded into the goals of providing technical, engineering and traffic analysis support to NSF High Performance Connections sites and the broad vBNS user community.

The NLANR Cooperative Agreement included tasking for the following areas:

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

  2. Testing and measurement of the vBNS performance characteristics.

  3. Coordination and oversight of the use of the vBNS as a shared facility among the supercomputing centers and selected Research and Education sites.

  4. Coordination and scheduling of utilization of the vBNS by researchers identified and referred to the SCCs, including the 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.

  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.

NLANR activities focused on 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 has served as the foundation for classes and qualities of service, accounting, and related economic models, and has enhanced the ability to engineer next-generation networks.

NLANR 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) were devoted to technical support. In the latter part of the NLANR project, the focus of participating sites became supporting the larger NSF high performance connections community and expanding related research. Site personnel continued to provide ongoing support related to technical and policy issues relating to vBNS connections.

Organizational Changes

Functionally, NLANR became divided into three parts:

During FY1997, NLANR leveraged the experience of the initial vBNS sites, providing a more proactive approach to supporting the vBNS users, particularly the 50+ institutions connected to the vBNS during FY1997 as part of the NSF Connections Program. NLANR also expanded its highly acclaimed research efforts relating to measurement, caching, and visualization.

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

  1. User Services / Outreach - user support, tracking and training was broadened to encompass 50+ NSF Connection Program sites connecting to the vBNS. The increase in support was addressed in large measure through a separate proposal from NCSA to NSF in Spring 1997, which became the Distributed Applications Support Team (DAST).
  2. Engineering - technical/operations support coordination. This area was led by Jamshid Madahvi at PSC. Engineering support encompassed coordination of technical discussions, i.e., planning, general meetings of new sites, and related meetings, such as those of NANOG, IETF, and Internet-2. NLANR's Engineering Working Group also monitored the performance of applications running over the vBNS and new connections research infrastructure. This group was also responsible for forging technical exchanges with CA*net2/NTN, DREN, ESNET, and other high-performance network infrastructure groups during FY97.

    Specific activities which were addressed by each of the participating NLANR sites included:

    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 assistance with the implementation of the NLANR Web caching hierarchy.

  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 continued to lead this topic area in 1997. Specific activities for FY1997 included: 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.

    In addition to these three Topic Areas, resources were also applied to overall management of NLANR and to provide NSF with Technical Feedback on Policy, as needed. This support was primarily provided by the Principal Investigators associated with each of the SCC sites.

vBNS

The very high performance Backbone Network Service (vBNS) is a nationwide network that supports high-performance, high-bandwidth research applications. The vBNS operates at a speed of 622 megabits per second (OC12) using MCI's network of advanced switching and fiber optic transmission technologies. At speeds of 622 megabits per second, 322 copies of a 300-page book can be sent every seven seconds.

Launched in April 1995, the vBNS is the product of a five-year cooperative agreement between MCI and the National Science Foundation (NSF) to provide a high bandwidth network for research applications.

The vBNS was designed for the scientific and research communities and originally provided high speed interconnection among NSF supercomputing centers and connection to NSF-specified Network Access Points. Today the vBNS connects two NSF supercomputing centers and research institutions that are selected under the NSF's high performance connections program.

The vBNS is only available for meritorious research projects with high bandwidth uses, and is not used for general Internet traffic.

Details on MCI/vBNS activities are covered in their monthly and quarterly reports for the vBNS Cooperative Agreement, available at www.vbns.net.

Highlights of MCI/vBNS activities in 1997 included:

High Performance Connection Sites

As the original Internet became congested with commercial and private traffic, the vBNS was created by the NSF and implemented by MCI Telecommunications Corporation as a high-performance data conduit among the nation's premier research and engineering organizations. The technologies developed for the vBNS are expected to lead to the "Internet of the future," as exemplified by the Clinton administration's "Next Generation Internet" initiative. Data transfer on the vBNS already is about 100 times as fast as on the commercial Internet, and soon will quadruple in speed.

In 4Q 97, PSC work on HPCS included:

In 4Q 97, NCAR work on HPCS included: In 1997 the U.S. National Science Foundation (NSF) released a solicitation, to prepare the groundwork for the development of the next generation of high-performance international research and education (R&E) Internet services (HPIIS). The goal of this solicitation is:
"to assist the U.S. research and education community (R&E) in meeting its needs for next generation international Internet services." . . . "HPIIS will seek high-performance connectivity between the NSF's very High Speed Backbone Network Service (vBNS) and high-performance networks of major international research partners."

A Current International Internet Access Point

One international access point to the vBNS has been established, namely the Science, Technology and Research Transit Access Point (STAR TAP).

Funded for three years by the National Science Foundation, the STAR TAP project will provide a persistent connection point for U.S. and international high-performance research networks and will improve the speed and performance of the applications that will run over these networks.

The STAR TAP -- Science, Technology And Research Transit Access Point -- is a persistent infrastructure, funded by the NSF CISE Networking and Communications Research and Infrastructure division, to facilitate the long-term interconnection and interoperability of advanced international networking in support of applications, performance measuring, and technology evaluations. The STAR TAP anchors the international vBNS connections program.

Physically, it connects with the Ameritech Network Access Point (NAP) in Chicago, as does the vBNS and other high-speed research networks. It enables traffic to flow to international collaborators from the approximately 100 U.S. leading-edge research universities and supercomputer centers that are now, or will be, attached to the vBNS or other high-performance U.S. research networks.

This project provides a persistent connection point for U.S. and international high-performance research networks, and its networking engineers will work to improve the speed and performance of the applications that run over these networks.

It is possible that other international access points to the vBNS will also be established.

In 4Q97, NCAR worked with Dr. Oliver McBryan (CU Primary Investigator) to review a joint vBNS application proposal between the University of Colorado and Germany's GMD to interconnect via the Canadian CANARIE (Star TAP) connection to the vBNS. The application itself includes: visualization, steering, distributed computing and collaboration tools in several network-based research projects.

The URL for this application is located at:

http://wwwmcb.cs.colorado.edu/home/mcbryan/cu-gmd-app.html

General Accomplishments

User Services

Applications supported by NCAR in 3Q 97 included:

Applications supported by PSC in 3Q 97 included:

Applications supported by PSC in 4Q 97 included:

Applications supported by NCSA in 4Q 97 included:

Applications supported by NCAR in 4Q 97 included:

Engineering Support

The NLANR Engineering web page developed by PSC can be found at NLANR Engineering . PSC has also established and maintained route lists for all primary and secondary vBNS sites. The route list allows HPC sites to determine what primary and secondary sites are accessible via the vBNS using network address, site name, or AS number to identify the site.

PSC continues to maintain a TCP performance tuning website at http://www.psc.edu/netwo rking/perf_tune.html.

Engineering Support activities for PSC for 4Q 97 included:

Engineering Support from NCAR in 4Q 97 included:

Engineering Support from NCSA in 4Q 97 included:

Research

NCSA Research work in 4Q 97 included:

Follow On Projects

NLANR Follow On Projects have been created for allthree areas of the NLANR program - Research and Measurement (CAIDA, NLANR/MOAT), Engineering (NCNE) and User Services (DAST).

CAIDA

The Cooperative Association for Internet Data Analysis (CAIDA) is a collaborative undertaking to promote greater cooperation in the engineering and maintenance of a robust, scalable global Internet infrastructure. CAIDA provides a neutral framework to support these cooperative endeavors.

The University of California, San Diego's (UCSD) proposal for seed monies to establish CAIDA was funded effective September 15, 1997. As described in the proposal , having the National Science Foundation complement an industry-focused effort with government support will promote balance among the needs of the various communities (private, research, government, and users) and facilitate the near term development and deployment of critical measurement technology and techniques.

CAIDA was originally established as a project of the National Laboratory for Applied Network Research (NLANR) within the University of California, San Diego. In May 1997, NLANR/CAIDA hosted the second in a series of Internet Statistic and Metrics Analysis (ISMA) workshops. During CAIDA's first year, staff will work with participating companies to define the goals, priorities, and desired membership of CAIDA. CAIDA's Program Plan for 1998 describes current activities. The program plan for 1999 will be developed cooperatively with CAIDA sponsors and members.

In July 1998, the Defense Advanced Research Project Agency (DARPA) awarded UCSD/CAIDA a $2.4 million award for a Next Generation Internet (NGI) initiative, see http://www.caida.org/NGI. This effort focuses on monitoring, depicting, and predicting traffic behavior on current and advanced networks, through developing and deploying tools to better engineer and operate networks and to identify traffic anomalies in real time. These efforts concentrate on: (1) developing tools to automate the discovery and visualization of Internet topology and peering relationships (through application of the CAIDA-developed Skitter tool); (2) monitoring and analyzing Internet traffic behavior on high speed links (through the development of OC48 and Light monitors); (3) detecting and mitigating threats (through security application of the Coral monitors); and (4) providing for storage and analysis of massive volumes of traffic data. This project leverages existing NSF-supported work, including efforts by the MCI/vBNS team, CAIDA and NLANR on the development and deployment of Coral monitors (OC3mon and OC12mon); deployment of the Skitter performance tool; and analysis/visualization techniques developed through NSF support of CAIDA. Key collaborators on this effort include MCI/vBNS, NTON, Abilene, and numerous networks and organizations hosting Skitter measurement hosts.

NLANR/MOAT

In May 98, the Measurement and Operations Analysis Team (MOAT) was established as a formal group within NLANR to analyze network traffic patterns and traffic behavior, evaluate service models, and conduct research to enhance the NSF's very high performance Backbone Network Service (vBNS).

Located on the UCSD campus at the San Diego Supercomputer Center (SDSC), the NLANR/MOAT group measures and analyzes vBNS traffic statistics to conduct network performance analysis and continues NLANR's efforts to promote the continuing development and deployment of related measurement and analysis tools and techniques.

NLANR/MOAT's principal activities include:

NCNE

The National Center for Network Engineering, based at the Pittsburgh Supercomputing Center, provides engineering services for the National Laboratory for Applied Network Research. NLANR's Engineering Services group provides in-depth information and technical support for connecting to and effectively using high performance wide area networks, such as the NSF's very high performance Backbone Network Service (vBNS) to campus network engineers, gigapop operators and other high performance networking professionals.

NCNE's web page can be found at http://www.ncne.org/

DAST project

In the fourth quarter of 1997, NCSA staff Charlie Catlett, Randy Butler, Ginny Hudak-David and Jim Ferguson worked on the establishment of the new NLANR follow-on Distributed Applications Support Team. The DAST web page can be found at http://dast.nlanr.net

The Distributed Application Support Team is a single point of assistance for vBNS users. DAST supports:

NLANR/Cache

Web caching has been popular internationally for several years due to the significant inter-continental bandwidth savings it provides. Only recently, however, has caching gained favor in the United States among commercial providers and large institutional users. In response to these trends, CAIDA and NLANR/Cache staff are collaborating with MAE-West and several participating ISPs (most notably Zocalo) to implement a prototype commercial cache model at the MAE-West Exchange Point. The collaboration began in December 1997. Unlike the successful NLANR Global Cache Hierarchy 'research' model, this new cache model requires participating ISPs to be responsible for their own http misses, e.g., provide their own transit to resolve http misses as opposed to reliance on the vBNS or other third party transit provider. The prototype, described at http://ircache.nlanr.net/Cache/mae-west/, now has 34 networks peering with the cache's router. Additional details on the NLANR-Cache effort are available at http://ircache.nlanr.net/. CAIDA and NLANR/Cache are also collaborating on the evaluation of vendor-specific cache solutions. As September 1998, two vendors are providing equipment to NLANR/CAIDA for testing, with equipment from a third expected by late 1998.

Star Tap project Star Tap is another project that resulted from the NLANR activities at NCSA. The activities and services of this project can be viewed at http://www.startap.net/

The STAR TAP (Science, Technology And Research Transit Access Point) is a persistent infrastructure, funded by the NSF CISE Networking and Communications Research and Infrastructure division, to facilitate the long-term interconnection and interoperability of advanced international networking in support of applications, performance measuring, and technology evaluations. The STAR TAP anchors the international vBNS connections program.

Physically, it connects with the Ameritech Network Access Point (NAP) in Chicago, as does the vBNS and other high-speed research networks. It enables traffic to flow to international collaborators from the approximately 100 U.S. leading-edge research universities and supercomputer centers that are now, or will be, attached to the vBNS or other high-performance U.S. research networks.

This project provides a persistent connection point for U.S. and international high-performance research networks, and its networking engineers will work to improve the speed and performance of the applications that run over these networks.

Attachments

Previous Reports

Past NLANR quarterly reports can be found at: http://oceana.nlanr.net/Reports/

Papers and Presentations

Past presentations from NLANR/CAIDA can be found at:
http://www.caida.org/Presentations/

Past papers from NLANR/CAIDA can be found at:
http://www.caida.org/Papers/

NLANR/MOAT presentations can be found at:
http://moat.nlanr.net/Presentations/Internet/