[e2e] [Fwd: State of MPLS deployments today]
avg at kotovnik.com
Sat Oct 5 01:30:48 PDT 2002
On Fri, 4 Oct 2002, Dennis Ferguson wrote:
> And while MPLS traffic engineering may be sloppy and imperfect, it does
> have the advantage that it only ever needs to compute a result for a
> single network topology, the one you currently have, and that is a
> substantial simplification.
Well... how much "optimal" traffic flows will be different from the
shortest-path flows even in case of single-link or single-router failures?
I strongly suspect that in most real networks this won't be too different.
I'm not aware of studies on determining a class of network topologies
which produce exactly the same results with and w/o MPLS - but muy gut
feeling that most providers have topologies which are pretty close, since
backbone layouts were selected way before MPLS became a reality.
Besides, backbone link failures are mostly handled by APS, and with some
trickery one can use APS for switching a backbone link between paired
routers in case if one goes down :)
> That said, I think most deployments of MPLS are not driven by traffic
> engineering issues, but by "how can I do more stuff with my network and
> make more money" speculative issues.
Absolutely :) MPLS VPNs are a bogosity, too.
More information about the end2end-interest