On Tuesday, October 25, 2011 08:37:39 AM Livio Zanol Puppim wrote: > Searching the Internet about this topic, I've found a > draft that seems to be a near future solution for some > major MPLS networks, called "seamless MPLS". As written > in the laymann draft ( > http://tools.ietf.org/id/draft-ietf-mpls-seamless-mpls-00 > .txt), the implementation uses 2 levels of IS-IS (or 2 > OSPF areas).
I looked at this draft a long time ago, and had a chat with the authors about some of the concepts. In general, the draft is fine, but a little complicated. It calls for confederations and BGP Labeled Unicast (RFC 3107). However, I do understand that if you're talking about a network of that size, it likely is a viable way to scale the interior routing domain. There are also complications with running multiple IS-IS levels when you're implementing RSVP-TE. For us, we're heavy on NG-MVPN with p2mp LSP's, so that could be problem. Of course, if you're doing MPLS-TE across IS-IS levels, expanded loose hops becomes your only option when you have a hierarchical IS-IS implementation. At any rate, we've now operationalized a full end-to-end IP/MPLS Metro-E network (thank God for no more spanning trees), which implements Seamless MPLS, but with the usual architecture and not the decentralized core, i.e., IS-IS for the IGP, iBGP and MPLS. This is on the ME3600X for the Access. I plan to discuss our deployment experiences at an industry event near you. I probably should get back in touch with the authors of Seamless MPLS, just to touch base :-). Mark.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
