[rrg] [Int-area] Please respond: Questions from the IESG asto whether a WG forming BOF is necessary for LISP

[Templin, Fred L]

Hi Dave, 

> -----Original Message-----
> From: David Meyer [mailto:dmm at 1-4-5.net] 
> Sent: Thursday, January 22, 2009 4:05 PM
> To: Yakov Rekhter
> Cc: int-area at ietf.org; rrg; routing-discussion at ietf.org
> Subject: Re: [rrg] [Int-area] Please respond: Questions from 
> the IESG asto whether a WG forming BOF is necessary for LISP
> 
> 	Yakov,
> 
> > In answering this question we need to keep in mind that such
> > techniques as (a) caching routing and forwarding information, (b)
> > use of separate mapping system for Loc/ID mapping, (c) relying on
> > probing for determining path feasibility are essential/fundamental
> > to LISP.
> 
> 	(c) is clearly not fundemental to LISP. 
> 
> 	What is fundemental (IMO) to any Loc/ID split
> 	architecture is what have what we have termed the Locator
> 	Path Liveness problem. How that problem is solved is a
> 	matter for each architecture to consider. 
> 
> 	In particular, probing is a solution to the Locator Path
> 	Liveness problem, but as we know from shim6 (etc), the
> 	complexity of that isn't pretty. But then either is the
> 	O(n!) complexity of BGP path hunting. In any event, the
> 	space of potential solutions for map-and-encap schemes
> 	such as LISP (or eFIT, Ivip, etc) haven't been throughly
> 	explored. This is part of the reason why we think LISP
> 	should be EXPERIMENTAL.

Not meaning to pick on you, but I don't understand why RANGER
hasn't made it onto the "short list" of proposals yet; maybe
you could answer that for me? AFAICT, RANGER is in the top-two
map/encaps proposals in terms of technical completeness (maybe
even the top-one) so can we have folks add it to their lists?

Thanks - Fred
fred.l.templin at boeing.com

> > In other words, these techniques form the foundation of
> > LISP.
> 
> 	With respect to LISP and probing, that statement is
> 	incorrect. 
> 
> 	Dave
>