Internet2 Engineering Update Guy Almes Internet2 Chief Engineer Educom Meeting Minneapolis — 30 October 1997
Outline of the Talk Internet2 Engineering Objectives Working Groups GigaPoP Progress Four Key Engineering Issues Large Delay-Bandwidth Products Introducing Quality of Service Improving Multicast Support Introducing IPv6
Internet2 Engineering Objectives Enable Advanced Applications Strengthen the Universities in their Research / Education Missions Pioneer Specific Technical Advances Establish GigaPoPs as Effective Service Points
Applications and Engineering Applications Engineering MotivateEnables
Comments on Apps and Plumbing Advanced applications transform high-speed plumbing into value Advanced plumbing enables advanced applications Profligate use of bandwidth, per se, does not make an application ‘advanced’ Megalomaniac plumbing, per se, does not make the plumbing ‘advanced’
Comments on the University Research/Education Mission Due to their teaching mission, universities scatter researchers University faculty and students therefore have a disproportionate need to be able to collaborate at a distance
Sketch of Internet2 Architecture Interconnect gigaPoP u u u u u u u u Interconnect: connects all the gigaPoPs to each other GigaPoPs: connect universities to the Interconnect and to other services Universities: upgrade their LANs to more than 500 Mb/s u u gigaPoP
1997 vs 1998 Sets of Aspirations 1997 High-speed uncongested best-efforts IPv4 T3 and OC3 will be typical; some OC12 About 15 gigaPoPs; about 45 universities Introduction of Measurements 1998 Introduce Quality of Service Improve Multicast Support Introduce IPv6
Working Groups to address project-wide technical issues minimal constraint on natural diversity of gigaPoP technical choices complementary to groups such as the IETF
Initial Working Groups IPv6: Dale Finkelson of Univ Nebraska Measurement: David Wasley of UCOP Multicast: Dave Meyer of Univ Oregon Network Mgmt: Mark Johnson of MCNC Quality of Service: Ben Teitelbaum (staff) Routing: Steve Corbato of Univ Washington Security: Peter Berger of Carnegie Mellon Topology: Paul Love (staff)
Oct-97
Nov-97
Dec-97
early 1998
Four Key Engineering Issues Large Delay-Bandwidth Products Introducing Quality of Service Improving Multicast Support Introducing IPv6
Large Delay-Bandwidth Products As the product of delay and bandwidth grows: The number of unacknowledged packets grows It becomes more difficult to sustain a steady stream of data from end to end Several consequences: Need for direct physical paths Tradeoff between buffering and variation in delay
Introducing Quality of Service Technical: End-to-end vs Intermediate Host vs Proxies Bandwidth, Delay parameters Administrative: Admission Control Measurements Authentication
Quality of Service Sketch Does the QoS approach support the applications? Are there implementations that work? Only one? If cloud ‘A’ and cloud ‘B’ both implement QoS, does the combined A+B catenation implement QoS? A B
Improving Multicast Support Current MBone community is small Many advanced applications are naturally multicast one to many (e.g., distance education) few to few (e.g., graduate seminars or conferences) Scaling is hard: Optimize for transmission lines? Optimize for packet forwarding?
IPv6 Issues Initially this will appear to be an end in itself We hope/expect that it will become an aid to solving other problems Compact Routing Tables Some help for QoS, IP options Products will be available beginning 1997