The MPLS WG Archive

[Yakov Rekhter <yakov@cisco.com>]

Jim,
 
> > Just to beat a dead horse .. the difference between the open model
> > and the overlay is just that the overlay model typically suggests that
> 
> That depends... see below.
> 
> > there would be N^2 connectivity (i.e., every router, for instance, is
> > statically connected to every other). In any case, the open model allows
> 
> While there may indeed be some misunderstanding as to the meaning of the
> "overlay model" (not to mention the meaning of the other models :-)), the
> traditional definition of overlay is not one of static operation.  NHRP and
> MPOA, for example, are overlay in my book (and most folks that I know).
>
> They most certainly imply non-static connections by defintion (i.e., use of
> SVCs).  This having been said, I believe that we can do one of two things
> here.  One, keep the current terms and potentially risk an ad nauseum
> inspection of the meaning of these terms; or two, come up with a term which
> is satisfactory to all.  In the second case we would have models:
> 1) X
> 2) Peer
> where X is the PC term for what we want to get at.  So my opine is that
> X="augmented routing model"/ARM model.
> 
> Do I hear a hmmmm???

I think you do. And it comes from you - it is your own hmmm.

Yakov.