Internet Info 1994 March/Internet Info CD-ROM (Walnut Creek) (March 1994).iso/inet/ietf/orad/orad-minutes-92nov.txt

Text File | 1993-02-17 | 6.0 KB | 196 lines

Editor's Note: Minutes received 12/7/92

CURRENT_MEETING_REPORT_



Reported by Gene Hastings/PSC

Minutes of the Operational Area Directorate (ORAD)

Recruiting of ORAD Members

What is expected of recruits?


   o Provide guidance as to what needs attention (what WGs need to be
     formed?)  (Example:  mbone coordination)

   o Provide guidance to working groups in other areas, e.g.  BGP
     Deployment and IPv7.  For example Network Management and SAAG
     explicitly assign people to working groups.

   o Document review.  Particularly early on, i.e., I-D, etc.  As things
     are going in the POISED Working Group, it looks like the direction
     is for the IAB to delegate more of its activity and responsibility
     to the IESG which will increase the need for area advisory groups
     like ORAD.
     Two kinds of review:

      -  All kind of Operations working group documents.
      -  Selected review of other area groups (like ROAD stuff, etc.).


Discussion

There is a need for an explicitly nominated ORAD membership as distinct
from the open ORAD meetings at the IETFs.  This closed group will be
responsible for the above listed topics.  It is for example not
necessary that those part of the ORAD personally have to review
documents but that they shall see to that such a review is made.

Current Operations Area working group Chairs could be part of ORAD but
this is not implicitly required.  There is necessary with a group of
people that have the interest and time to undertake the ORAD
responsibilities.

There is a need for a method of flagging documents for ORAD review.  If
enough ORAD members thinks it needs ORAD review one member is assigned
the responsibility to see that this happen.

To make ORAD having broad coverage it will be necessary to invite
operators that traditionally do not participate at IETFs.

Interested in ORAD participation:

                                   1





Tony Bates          University of London
Nevil Brownlee      University of Auckland
Henry Clark         OARnet
Michael Conn        MCI
John Curran         NEARnet
Phill Gross         Advanced Network and Services, Inc.
Daniel Karrenberg   RIPE NCC
Peter Lothberg      EBONE
Bill Manning        SESQUINET
Bernhard Stockman   SUNET
Evan Wetstone       SESQUINET
Christopher Wheeler   University of Washington


Proposed Charter


  1. What is the ORAD (and what is it not).
  2. Forum for OPs groups.
  3. Development of methods and practices.
  4. Guidance and review.
  5. OPs information and education.


The need for a backbone requirements document was discussed.  There is
value to documents outlining needs, services, and interoperation, but if
it is too proscriptive, they may fail to accommodate all economic or
organizational models.

The meeting concluded that ORAD should not start off too big but
initially concentrate on document review and presentation of issues to
working groups.

Discussion around MBONE Coordination

There is a need to increase multicast performance in today routers

Matt Mathis volunteered to track MBONE contacts for the subversive
purpose of collapsing connections to the highest level possible.  The
right thing to do is prevent the mbone from being heavily used until
mrouted is fixed.  If the operators were to turn it off, however, there
would be a grass roots mbone appearing which we would have NO control
over.

                                   2





ORAD should issue a statement of recommendations on mbone utilization,
requirements and operation.  In the meantime, can we get debugging
tools, can we get multicast support from vendors?

Architectural weakness:  25 speakers at once fills a T1.  This would
create a situation of denial of service.

Mrouted needs more knobs.  Must be able to do route pruning.

Action items for mbone:


   o Major mrouted work
   o Get together and list some bullets to take to Steve Deering and
     Steve Casner.
   o Remove redundant tunnels.
   o Public versions released as receive-only?
   o Restrict to one audio and one video until more experiences.
   o More tests and freeze of topology.


Tests shall happen before IETFs which includes announcements of
tunneling and requests to be made further in advance of conferences.
Strict cut-off date after which no more tunnels

Others actions:


   o Need for more efficient diagnostic tools.
   o mrouted related work.

      -  Put throttling in the tunnels.
      -  Treatment for misconfiguration (view others' configurations).
      -  Pruning of the tree (no more than 12?).
      -  Encaps, not LSRR.
      -  Experiment with one-way path.
      -  Encourage codings which conserve bandwidth.
      -  Experiment outside of IETF meetings.


The Working Group needed to flesh these out with representatives from
Merit, PSC, NEARnet.

Attendees

Tony Bates               [e-mail]
Rebecca Bostwick         [e-mail]
J. Nevil Brownlee        [e-mail]
Henry Clark              [e-mail]
Michael Conn             [e-mail]
John Curran              [e-mail]
Hans Eriksson            [e-mail]
Dennis Ferguson          [e-mail]

                                   3





Richard Fisher           [e-mail]
Peter Ford               [e-mail]
Phillip Gross            [e-mail]
Robert Gutierrez         [e-mail]
Eugene Hastings          [e-mail]
Alisa Hata               [e-mail]
Daniel Karrenberg        [e-mail]
Mark Knopper             [e-mail]
Daniel Long              [e-mail]
Kim Long                 [e-mail]
Bill Manning             [e-mail]
Dennis Morris            [e-mail]
David O'Leary            [e-mail]
Andrew Partan            [e-mail]
Marsha Perrott           [e-mail]
Bernhard Stockman        [e-mail]
Marten Terpstra          [e-mail]
Evan Wetstone            [e-mail]
Chris Wheeler            [e-mail]
Paul Zawada              [e-mail]



                                   4