On 22 Apr 2010, at 23:22 , Robin Whittle wrote: > If so, how could this be true since a > dual-homed network would chew 3 times the amount of global unicast > space it actually uses?
It is NOT true that a multi-homed host needs more than 1 prefix with ILNP. Please re-read the 2009 MILCOM paper for a detailed explanation of one possible way that a site can be multi-homed with minimal consumption of global-scope locators/addresses. If one deploys the approach in that paper, it actually permits (at least in theory) some IPv4 address space to be reclaimed, as compared with today's BGP multi-homing, because only exterior interfaces of site border routers need to have global-scope locators/addresses. (In practice, the monetary value of already delegated IPv4 prefixes is probably too high for most organisations to volunteer to return them, but that is an economics issue rather than a technical issue.) > In that correspondence I didn't mention that ILNP wouldn't be useful > for IPv4 because I considered this to be an obvious fact which was > not in dispute. Several of us have repeatedly tried to explain to you that the above was not correct; we've tried both on-list and off-list. (I had thought this was NOT a debating society.) There is a difference between (A) one person disagreeing (or not understanding) part of a proposal and (B) the proposal actually having the issue that person claims. This topic seems to be (A). Yours, Ran _______________________________________________ rrg mailing list rrg@irtf.org http://www.irtf.org/mailman/listinfo/rrg
