Reason to use CEF in MPLS

Unanswered Question
May 23rd, 2007

Hi all,

Why only CEF is used in MPLS?

As far as I know CEF is the only switching process in Cisco IOS that can label an incoming packet and forward it. Any other reason.



I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
swaroop.potdar Wed, 05/23/2007 - 13:53

CEF is not only used in MPLS ,but rather its also used for MPLS.

Traditionally CEF has been used in IP routers to make forwarding decisions. What this means is routing tables

create RIB based on the routing information they have. This information is used by CEF to create a forwarding information

base (FIB) which has the outgoing interface information as well.

In addition to this it creates a Adjacency table, which is nothing but layer 2 addressing information to reach the next-hop IP on a outgoing interface. Using these two information the processor does express switching/forwarding.

(This enables the router not to do multiple lookups i:e: for destinationIP, then the outgoing interface and then the

Framing type on that outgoing interface)

Now, in MPLS since the router has to process Labels in addition to the Layer 2 encapsulation/decapsulation.

Its makes more sense and easier for the router to incorporate this Label rewrite information into the LFIB.

And make the forwarding decisions using the combination of LFIB and Adjacency table.

To summarize it, CEF is a forwarding mechanism based on the principle of seperation of control and forwarding plane to achieve fast switching. Different such mechnisms may be implemented by other vendors to do forwarding.

You can learn more about CEF here on this link.




This Discussion