Full Abstract

This tutorial introduces the Internet Routing Registry (IRR), the Routing Policy Specification Language, (RPSL) and the RtConfig router configuration tool. We explain how to register and query routing policy objects in the IRR. After a brief introduction to routing policies, we discuss RPSL, the language for specifying Internet routing policies in the IRR. We explain several policy examples currently practiced in the Internet and specify them using RPSL. These include as-path prepending, setting preferences and other metrics based on the community attribute, and access lists based on prefix and as-path expressions.

RtConfig is the focus of this tutorial. RtConfig can configure Cisco, Juniper, Nortel, Gated and RSd routers with BGP policies. The amount of configuration it performs is controllable; hence, the amount of information that needs to be published in the IRR is controllable as well. RtConfig can be used to only generate prefix access lists from the IRR, or to generate full-blown router configurations.

Speakers
Cengiz Alaettinoglu, ISI
Cengiz Alaettinoglu works on the Routing Arbiter project at the USC Information Sciences Institute. He co-chairs the IETF Routing Policy System WG, which defined the Routing Policy Specification Language along with the protocols to enable a distributed, secure routing policy system.

Full Abstract

Speakers
Steve Deering, Cisco Systems
Steve Deering is a Technical Leader at Cisco Systems , where he is working on the development and standardization of architectural enhancements to the Internet Protocol. Prior to joining Cisco in 1996, he spent six years at Xerox's Palo Alto Research Center, engaged in research on advanced internet technologies, including multicast routing, mobile internetworking, scalable addressing, and support for multimedia applications over the Internet. He is a member of the Internet Architecture Board, a present or past chair of numerous Working Groups of the Internet Engineering Task Force (IETF), the inventor of IP Multicast, and the lead designer of the new version of the Internet Protocol, IPv6. He received his Ph.D. from Stanford University.

Full Abstract

A half-day, interactive tutorial that discusses and demonstrates the interaction between MPLS and the traditional IP routing protocols. The tutorial begins by examining the general requirements for traffic engineering and then presents how specific features of MPLS address these requirements. The tutorial then proceeds to address how advanced features of MPLS work in conjunction with BGP and the IGP (IS-IS or OSPF) to control network traffic.

During the tutorial, the presenters will augment the lecture material by demonstrating various MPLS features including the signaling of LSPs with RSVP, mapping routes to LSPs, configuring LSPs for transit and local traffic, and enabling advanced traffic protection features.

Topics to be covered include:

• Signaling LSPs with RSVP
• Constraint-based traffic engineering
• Extensions to IS-IS and OSPF
• Traffic Engineering Database
• Strict and loose routing
• Administrative groups (a.k.a. Colors)
• Hot-standby and Fast Reroute LSPs
• Applications and advanced features

Speakers
Jeff Doyle, Juniper
Chris Summers, Juniper Networks

Full Abstract

The purpose of this tutorial is to present a framework for the construction of a BGP architecture for mid- to large-sized ISPs utilizing BGP confederations. The primary building block for this approach is the concept of the "POP as Confederation". While concentrating on theory, this session will also highlight the real world implementation of this approach on Cisco routers at Mindsping, and on Juniper routers at NetRail. Elements include:

• Meaningful MEDs for customers, as well as basic, non-MPLS traffic engineering.
  
  
• A well thought-out community structure to optimize route filtering and propagation.
  
  
• Special areas of interest include supporting wholesale dialup and broadband providers (OPN) and allowing customers the maximum possible control of their routing.
  
  

Speakers
Miguel Dimayuga, Earthlink
Dan Golding, NetRail

Full Abstract

Penetration Rate of Private Address Space

Current Use of IPv6 Address Space

ARIN IPv6 Delegation Status

Speakers
Bill Manning, ISI.
Michael O'Neill, ARIN

Full Abstract

Speakers
Moderator - Bob Fink, Lawrence Berkeley National Lab
Bob Fink is the Associate Department Head for Research for the Energy Sciences Network (ESnet) at Lawrence Berkeley National Laboratory. ESnet provides Internet services and advanced technologies to the US Dept. of Energy (DoE)-funded national research laboratories. Bob is a co-chair of the ngtrans IPv6 Transition working group of the IETF, and has led the 6bone IPv6 Testbed project since its inception. He also formed the international 6REN initiative to create worldwide IPv6 production network services by organizing Research & Education Networks worldwide, and led the project to create the first USA located-IPv6 peering point for production IPv6 networks (the 6TAP).

Panelist - Bill Maton, Communications Research Centre
Panelist - Sean Mentzer, Qwest
Panelist - Greg Miller, MCI WorldCom
Greg Miller is the Director of Advanced Internet Technology at WorldCom. Greg's group is responsible for engineering and operating the vBNS network and for developing advanced IP services. In 1998, Greg designed and implemented a nationwide native IPv6 service on the vBNS backbone.

Panelist - Rob Rockell, Sprint
As a backbone engineer with Sprintlink, Rob Rockell deployed IPv6 first as a toy, and second as a means to provide more valued service to the research-focused segment of the Sprint customer base. Testing has been done with Cisco, Bay, Solaris, Linux, and Microsoft platforms.

Full Abstract

ZipLink is currently working on a technical solution to the problem of outbound spam. We are a wholesale dialup ISP, and as such, can be a source of spam. Our idea is to track, utilizing RADIUS accounting records fed into an Oracle database, the IP addresses and usernames of all customers logging into our dial up ports. Then, when a customer attempts to send email, their outbound mail connection is redirected to a mail relay which checks their current number of outbound messages against preset limits in the database. If they exceed any of the limits, the message is bounced/dropped. If they do not exceed the limit, the message is allowed to go through and their email counts in the database are updated to reflect their most recent activity.
We have successfully prototyped this functionality in a lab environment, and are just now beginning limited POP testing before a full scale rollout to follow later this year.

Speakers
Robert Haskins, ZipLink

Full Abstract

Speakers
Susan R. Harris, Merit Network

Full Abstract

An overview of recent IETF work to add security to the Domain Name System. Topics covered include DNSSEC (RFC 2535), Query/Response security, using DNS as a server of security data, and securing dynamic update. In addition, material on how DNS security will impact operation is included.

Speakers
Edward Lewis, TISlabs

Full Abstract

Speakers
Stefan Savage, University of Washington

Full Abstract

The CAIDA Metrics Working Group has produced a draft document, 'Network Measurement FAQ,' intended for people who buy Internet service and make it available to others, e.g. corporate users and smaller ISPs. This presentation will introduce the FAQ, highlighting its section on 'the Most Common Metrics.'

Common Metrics such as latency, packet loss and availability often appear in service desccriptions published on the Web, but they are seldom clearly stated. The Working Group's definition of availability, and methods of measuring it, will be explained.

Speakers
Nevil Brownlee, CAIDA

Full Abstract

This talk discusses the impact of the CALEA law on network operators. It covers such topics as:

• What is CALEA?
• The difference between CALEA and wiretapping.
• Definition of terminology used in CALEA
• What is a safe harbor standard?
• Compliance requirements on Service Providers and equipment vendors.
• Timeframes for compliance.
• Industry efforts

Speakers
Chip Sharp, Cisco Systems