The MPLS WG Archive

[Giles Heron <giles@packetexchange.net>]

On Fri, 2002-07-26 at 15:14, Shahram Davari wrote:
> Giles,
> 
> > 
> > I want *both* mp2p LDP and targetted LDP.
> > 
> > I want the former because it:
> > 
> > 1) minimises the number of labels required in the core - O(N) rather
> > than O(N squared).
> > 
> 
> I think Heinrich Hummel demonstrated in the Japan (as predicted by Dave Allan)
> that mp2p (LDP) results only ~15% less states than N^2 mesh. 

I didn't see Henrich's demonstration :(

>From what I remember he relies on an assumption of per interface label
state to achieve this (i.e. consider label L received on interface X and
forwarded to interface Z to be a different state to label L received on
interface Y and forwarded to interface Z).  This assumption is correct
for some MPLS P devices - but not AFAIK for those which are deployed
most commonly.

And of course the results will vary depending on the ratio of P to PE
devices and the on the network topology...

Giles

-- 
=================================================================
Giles Heron    Principal Network Architect    PacketExchange Ltd.
ph: +44 7880 506185              "if you build it they will yawn"
=================================================================