NGI Implementation Plan
Section 2.2.3: Specifics/NSF
leftright
2.  Goal 1: Experimental Research for Advanced Network Technologies
        (continued)
2.2  Agency Specifics (continued)
2.2.4  National Science Foundation
2.2.4.1  Introduction
2.2.4.2  Nextwork Growth Engineering
2.2.4.3  End-to-End Quality of Service
2.2.4.4  Security
2.2.4.5  Milestones


2.  Goal 1

Goal 1: Experimental Research for Advanced Network Technologies (continued)



2.2
Agency Specifics


Agency Specifics (continued)



2.2.4
NSF


National Science Foundation



2.2.4.1
Introduction

Introduction
 
NSF projects make a central contribution to NGI goals by leveraging extensive campus and industry partnerships to connect about 100 leading research universities with a high performance network fabric, interconnecting this fabric with that of other Federal and foreign research networks, implementing and testing advanced technologies, and supporting hundreds of advanced applications. The present goals of the NSF programs for high performance connections reflect all aspects of the NGI: high performance connections to about 100 universities and their research partners with QoS technology supporting hundreds of meritorious applications.
 
NSF has an experienced and capable research base on which to build its existing high performance connections to the NSF supercomputer centers and their partners (Partnerships for Advanced Computational Infrastructure, PACI), the NSF-sponsored National Laboratory for Applied Networking Research (NLANR), and dozens of funded individual investigators in university and industry laboratories, as well as ongoing funded research with investigators in DARPA, NASA, and other agencies. Moreover, NSF's partnership with Internet2 will focus the collective efforts of over 100 leading universities on next generation network technologies and related issues of deployment, management, and testing for NGI Goal 1.



2.2.4.2
Network Growth
Engineering

Network Growth Engineering
 
The goals of this task are to (1) create and deploy tools and algorithms for planning and operations that guarantee predictable end-to-end performance at scales and complexities of 100 times those of the current Internet; (2) facilitate management of large scale internetworks operating at terabit speeds supporting a range of traffic classes on a shared infrastructure; and (3) create an infrastructure partnership in which lead users (government and research) share facilities with the general public, thereby accelerating the development and penetration of novel network applications.
 
Planning and Simulation
 
NSF will work with approximately 100 leading universities and the Internet2 organization on policies and technologies for planning large scale, high performance networks. Internet2 and its members will be coordinating 15 to 20 GigaPOPs linking institutional, state, and regional high performance networks into a national system that must support a wide variety of new technologies and applications.
 
Monitoring, Control, and Display
 
NSF and its National Laboratory for Applied Networking Research are leaders in the deployment of network measuring and monitoring devices and in the theoretical analysis of network behaviors. NSF is also funding related measurement efforts by Merit, the Common Solutions Group, and others. All will be coordinated to focus on the study of high performance networks.
 
Integration
 
NSF will work with DARPA and NASA to support the integration of agency research networks and the sharing, where possible, of leading edge research and robust production infrastructure. Integration will be a major activity and interest at the 15 to 20 Internet2 GigaPOPs (Points of Presence capable of network interconnections at gigabit speeds).
 
Data Delivery
 
NSF's NLANR will extend its on-line, real time tool set for the visualization of network configuration and performance, focusing on the complexities of integrated networks that are of mixed technologies.
 
Managing Lead user Infrastructure
 
Lead user management strategies and technologies will be a central focus of the NSF Partnerships for Advanced Computing Infrastructure (PACI). Both the San Diego Supercomputing center (SDSC) and the National center for Supercomputing Applications (NCSA) science will depend heavily on successful approaches to these problems, and both PACI and the Internet2 universities will devote considerable expertise and effort (over 200 lead network managers and technologists) toward practical, common solutions.



2.2.4.3
End-to-End Quality
of Service

End-to-End Quality of Service
 
The goals of this task are to facilitate the delivery of end-to-end assured QoS to applications and to ensure that these technologies can be tailored for use by lead users who have demanding requirements.
 
Baseline Quality of Service Architecture
 
The baseline QoS management architecture will provide the framework of models, languages, and protocols to permit distributed applications to specify multidimensional QoS requirements, to negotiate acceptable trade-offs and confidence levels, and to receive feedback on delivered QoS enabling adaptation. NSF and its Internet2 partners will select, test, and deploy technologies for QoS on the vBNS and connected campus networks at an accelerated schedule. QoS related activities are perhaps the major technology goal of Internet2, which will be exploring multiple solutions and their interoperation at 15 to 20 GigaPOPs and at about 100 campuses in addition to the national networks. NSF will coordinate these activities closely with DARPA and NASA to help achieve an early, interoperable implementation.
 
Drill down Technologies
 
NSF's vBNS presently supports simple drill down technologies such as cut through routing. This early effort will be dramatically expanded to encompass multiple approaches at the Internet2 GigaPOPs and regional networks, and will be broadened to include the development of corresponding measurement and management tools.
 
Next Generation Network Technologies
 
NSF's Internet2 and vBNS partners are committed to early adoption of next generation technologies such as IPv6, RSVP. These technologies will be installed and operated directly on the vBNS and other Internet2 networks as soon as testing and management issues permit. NSF will extend the existing NLANR to include the direct participation of about 100 Internet2 members, and will work cooperatively with them as well as with its supercomputer center sites on national scale implementation and testing of new technologies and services such as:

  1. QoS trials in IP and ATM
  2. Performance measurement and statistics
  3. Caching, multicast, virtual testbeds
  4. End-to-end performance tuning for systems and applications
  5. Support role in universities for applications development
In addition, NSF will work with other agencies through joint solicitations to support individual R&D in network technologies including: QoS, Multicast, IPv6, scalability, security, and others.



2.2.4.4
Security

Security
 
The NGI security goals are to provide the basis for implementing and enforcing appropriate security policies among organizations, users, and infrastructure components under shared control.
 
NSF's primary activities related to security will consist of early implementation and testing of security technologies and policies selected in cooperation with its Internet2 and vBNS partners. NSF will coordinate these activities for interoperability with NSA and the other Federal agencies.



2.2.4.5
Milestones

Milestones

  FY1998 (1Q)   Extend the NLANR to include technical staff of all connected Internet 2 sites
  FY1998 (2Q) Make joint announcements with other agencies for research awards in network technologies
  FY1998 (3Q) Provide integrated IP and ATM monitoring and analysis tools at all sites
  FY1998 (3Q) Play a formal NLANR support role in universities for applications development
  FY1998 (4Q) Implement caching and native multicast at all appropriate sites
  FY1998 (4Q) Make awards to campus PIs for new research projects
  FY1999 (1Q) Implement RSVP for weighted early drop or other QoS, etc.
  FY1999 (3Q) Implement security at all appropriate sites
  FY1999 (4Q) Implement more complex RSVP and additional services
 

leftright