On Thu, Jul 10, 2008 at 1:59 PM, Tony Li <[EMAIL PROTECTED]> wrote: > MPLS requires that you install state _per tunnel_ on intermediate > forwarders. In other packet based tunnelling protocols, this is not > necessary. > > Both MPLS and normal L3 protocols require routing state on intermediate > forwarders.
Hi Tony, I'm still not following the distinction. I thought the way MPLS worked was along these lines: Router A announces, "I know what to do with label 555." This label is distributed to the rest of the systems speaking MPLS. Any packets they see where the top label is 555, they forward them in the direction they consider optimal to reach router A. Router A then speaks a routing protocol (e.g. BGP) to router B. There are multiple MPLS routers in between the two but routers A and B are speaking BGP directly to each other. Router A tells router B: "Any IPv4 packets you want to send to me, label them 555 and move them through MPLS." As a result, router B can move packets to router A without the intervening routers needing the full BGP table. They only need to know label 555. But that's map-encap in a nutshell. You map multiple discontiguous address spans into a single label that represents those spans' shared exit point. You then route through the intermediate (core backbone) nodes on the (presumably) much smaller number of labels. Whether that label is part of a completely different protocol (MPLS) or just a new address in the same protocol, the map-destination-to-a-label, encapsulate and route-on-the-label process is semantically the same. Have I grossly misunderstood how MPLS works? What's the story? -Bill -- William D. Herrin ................ [EMAIL PROTECTED] [EMAIL PROTECTED] 3005 Crane Dr. ...................... Web: <http://bill.herrin.us/> Falls Church, VA 22042-3004 -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
