Internet Architecture Board


IAB Minutes 2009-01-28

Home»Documents»Minutes»Minutes 2009»IAB Minutes 2009-01-28

IAB Teleconference


1. Roll-call, agenda-bash, approval of minutes, administrivia

1.1. Agenda

1.2. Attendance

  Loa Anderson (IAB Liaison to the IESG)
  Gonzalo Camarillo
  Stuart Cheshire
  Russ Housley (IETF Chair)
  Olaf Kolkman (IAB Chair)
  Gregory Lebovitz
  Barry Leiba
  Kurtis Lindqvist
  Andy Malis
  Danny McPherson
  Lynn St. Amour (ISOC Liaison)
  Dow Street (IAB Executive Director)
  Dave Thaler
  Lixia Zhang
  Lars Eggert (IESG Liaison to the IAB)
  Aaron Falk (IRTF Chair)
  Sandy Ginoza (RFC Editor Liaison)
  David Oran

1.3. Administrivia

The group had planned to review the new IESG slate, but deferred the discussion because the revised slate was not yet available from the NOMCOM.

1.4. Meeting Minutes

No meeting minutes were reviewed during this meeting.

2. Headers and Boilerplates

Olaf reported that he had received a few substantive questions and several nits on the Headers and Boilerplates document. He will draft a version -07, and asked the IAB to be ready for a short- turnaround review. It was expected that the -07 draft would be submitted for publication, with any remaining nits handled via the RFC Editor. There will also be one more week of review on the RFC-interest mailing list.

3. IETF 74 Plenary

The group continued coordination on the MPLS plenary topic. There was agreement that the session should focus on lessons learned from MPLS development and deployment, particularly those that might be applicable to other IETF work. It was decided not to focus on inter-SDO coordination, even though this aspect of MPLS has entailed significant work.

There was further discussion about how to frame the session in terms of “What Makes for a Successful Protocol”. Loa suggested something like (1) a ten year perspective, (2) an operator perspective, and (3) a talk along the lines of “what happened in the market”. Kireeti Kompella and George Swallow were identified as likely presenters; a speaker from the operator community still needs to be identified. Loa and Andy will finalize the list of speakers and contact them this week. The plan is to request draft slides from the presenters early enough to allow IAB review and possible iteration. This should help to ensure a cohesive session and avoid overlap.

4. IETF 75 Plenary

The IAB has been considering the topic of network neutrality for the IETF 75 technical plenary. Barry drafted some initial notes on how to construct a session on the topic, and the group discussed possible goals and “non-goals” of such a plenary. Based on a number of side conversations, there seems to be widespread interest both inside and outside the IETF for a plenary on this topic.

Several technical, political, and business aspects of the topic were considered, and there was consensus that simply providing education on the topic would be of significant benefit to the community. Olaf noted that Patrik Faltstrom has experience in this area, and suggested inviting him to assist the IAB with planning. Barry and Danny will lead the planning effort, and all IAB members were asked to review the material distributed so far.

5. NAT66

The NAT66 discussion picked up where it left off at the last meeting. Dave Thaler summarized several related issues of topology hiding and host hiding. Andy noted the interest in PI space, and how providers want as little PI space as possible (since the use of PI space often limits aggregation). Lixia commented that when the community asks for architectural guidance, the IAB should try and describe the architectural trade-offs, not simply agree or disagree. Andy raised the prospect that translation could occur at the edge of the network or deeper within the network infrastructure. This led to a discussion of symmetry – i.e., if one side deploys NAT, does the other side need to deploy it as well? Or does the deployment in one part of the network drive a new cost for other network users, even if they don’t want or need NAT?

There was some agreement that certain translation schemes might be designed so as to be transparent to other network users. It was also agreed that more information was needed about the perceived interest in topology hiding.

Lixia had prepared some slides on NAT66 as well, but the presentation was deferred due to time constraints. Lixia and Dave suggested three areas for solutions (as something for the IAB to think about until next time):

- use PI space everywhere (i.e. don’t renumber)
- use tunnels to the host, and renumber outside but not inside
- deploy a form of translation

The group will resume discussion of the topic at the 2/4 meeting.