[nsp] VPLS service on Cisco routers

[sthaug at nethelp.no]

> > > Well , one can say that Juniper is compatible only with itself ( as long
> > > as VPLS is concerned ) so it is not a big suprise :).
> > > Unless they would also implement Lasserre-V.Kompella draft there will be
> > > situation like Juniper VPLS vs Rest_of_the_world VPLS.
> >
> > That depends on what the customers demand. If enough customers find that
> > they prefer the Juniper implementation, even Cisco will change :-)
> >
> > We're going for the Juniper variant in a mixed Cisco/Juniper network
> > here.
> 
> As far as I understand the only difference about autodiscovery and
> signaling. That should be resolved at IETF an ask router vendors to
> implement the standard solution.

It may well be resolved by IETF - but some of us don't have time to wait
for that. This means that if we (the customers) want interoperable VPLS
implementations *today*, we need to demand it from our vendors.

Remember that new technology doesn't *have* to be in the form of an RFC
in order to be useful. An excellent example of this is Martini tunnels,
in wide use for L2 transport across an MPLS network.

Steinar Haug, Nethelp consulting, sthaug at nethelp.no