[c-nsp] MPLS questions

nasir.shaikh at bt.com nasir.shaikh at bt.com
Fri Nov 17 12:40:54 EST 2006



Hi Oli,
Thanks for the reply.

What I meant was that I would need to implement AToM if I needed EoMPLS
along with the L3 VPNs. I guess my query has been answered.

Regards

nash

-----Original Message-----
From: Oliver Boehmer (oboehmer) [mailto:oboehmer at cisco.com] 
Sent: vrijdag 17 november 2006 18:29
To: Shaikh,N,Nasir,JPEC R; cisco-nsp at puck.nether.net
Subject: RE: [c-nsp] MPLS questions

cisco-nsp-bounces at puck.nether.net <> wrote on Friday, November 17, 2006
6:02 PM:

  
> The current setup has dual 6504s on the sites and dual 6509 /
> sup720s a
> the HQ locations. There is direct fibre connection between
> them and I am
> running OSPF on the network (there are other sites connecting to HQ
> also). The idea is to implement MPLS at the HQ site and
> vrf-lite on the
> remote sites to support separate divisions of the same customer at the
> same site. As this is not a standard CE>PE>P setup, I am a
> bit confused.
> Should I consider the 6509s as the collapsed MPLS backbone. Am I then
> correct in assuming that there will not be actually any label
> swapping? 

Well, the 6509 in the HQ will act as PE device as well as P (for traffic
between the remote-site). No problem with that. If Site A or Site B
communicate with the HQ  , the only label you'll have on the frame is
the VPN label (or VC label if you use EoMPLS).
 
> Can I also have EoMPLS between siteA and siteB over the same
> infrastructure (there is a single fibre between the sites and HQ)?

Yes, of course.. In this scenario, the HQ 6509 acts as P device.

> Is AToM the answer to the questions?

To which question? ;-) EoMPLS is a subset of AToM..

	oli



More information about the cisco-nsp mailing list