To anyone who has concerns about the prospect of Ivip being
chosen as the most promising proposal for solving the routing
scaling problem:
Please collaborate on a 500 word analysis / critique
for the RRG Report. I understand from Lixia's
msg05558.html that the deadline is Friday 15 January.
If possible, please read read the new Ivip-arch-03 ID:
http://tools.ietf.org/html/draft-whittle-ivip-arch-03
The Ivip page links to the IDs and has summaries of 10k,
2k and 1k words and a slightly updated version of the
TTR Mobility paper:
http://www.firstpr.com.au/ip/ivip/#docs
I suggest that anyone who wants to do this mention it
on the list and maybe send draft pieces of critique
to the list, as a way of encouraging others to add to
it or improve on it. I won't respond to these unless
I think they are based on a serious misunderstanding.
I am hoping that the analysis / critique will have something
positive to say about Ivip. However it is more important
that contains the major objections *anyone* has to Ivip in as
much detail as possible.
Quite a few RRG members have objections to any core-edge
separation scheme, because they believe core-edge elimination
is the way forward. As I wrote:
http://www.ietf.org/mail-archive/web/rrg/current/msg05562.html
there are 6 proposals for core-edge elimination schemes:
GLI-Split, hIPv4, ILNP, Name-Based Sockets, Name overlay (NOL)
and RANGI.
Another entry, which isn't really a proposal to solve the
scaling problem: "Aggregation with Increasing Scopes" suggests
that the long-term solution should be host-based - which
probably means core-edge elimination rather than core-edge separation.
I think these general objections to core-edge separation
schemes should part of the critiques of Ivip and likewise of
LISP and TIDR.
LISP folks - you must have serious objections to Ivip,
otherwise you wouldn't propose LISP-ALT. It would be great
if you would return the favour of all my critiques of LISP,
read the new Ivip-arch ID and contribute your criticisms.
If you don't have time for this, I guess you consider full-
database local query servers and/or pushing mapping around
the world in a few seconds are impossible or undesirable
- so please contribute those objections in as much detail as
possible.
I am happy to discuss any questions or concerns about Ivip -
ideally on the list, or privately by email or phone.
- Robin
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
