The field may have been well plowed by NIMROD, but the IETF forgot to water it. This organization has never sufficiently answered the route scaling problem, and the ISPs are paying for it today. The question is really whether IPv6 is properly deployable over the long term without a new multihoming paradigm (for `pick your definition of multihoming`).
The usual suspects (including Noel) are working through these scaling issues on multi6.
Eliot
Bill Manning wrote:
% The multi6 wg is working on scalable multihoming in IPv6. It looks like % this will be done by separating identifiers from locators. One (radical) % way to do this would be to use the full host name as the identifier and % IP(v6) addresses as the locators. Other than the fact that it breaks % all known transport protocols, this makes a lot of sense IMO. % % Iljitsch van Beijnumfor those doomed to about to repeat history... check the archives for NIMROD. or current work in DCCP. this is not greenfield, its well plowed ground. have you done your homework?
