Design Q: under 50 ms convergence time for large SP

From: jeremy co <jeremy.cool14_at_gmail.com>
Date: Tue, 26 Jun 2012 11:09:57 +1000

Hi,

We have a requirement of achieving under 50ms convergence time in large SP
network consist of lots of TE and PWs ,VPLS.

FRR implementation can achieve under 50 ms convergence time on TEs with
BFD. However, it cannot implemented in a large scale as it has to be
provisioned with each single node which requires a lots of work;meaning
NNHOP tunnel for protecting every node every link that terminates on that
node. each node terminating between 2 to 4 links. having around 80 nodes.
it will be roughly 300 NNHOP tunnels!!!! to cover every scenario.

Secondary option under TE tunnel will give us a good back up but it does
reduce the convergence time to in best scenario 70ms instead of 50 ms.

Appreciate any thoughts, experience

Thanks.

Jeremy

Blogs and organic groups at http://www.ccie.net
Received on Tue Jun 26 2012 - 11:09:57 ART

This archive was generated by hypermail 2.2.0 : Sun Jul 01 2012 - 10:39:52 ART