Hi,

> I think one has to distinguish where the bottleneck is:
> 1) link bandwidth (link congestion)
> 2) routing message processing (CPU congestion)

Actually it is much more complex than that.

The most often bottlenecks I have seen were either narrow channel into
RE/RP blades at a given router or  drop at ingress when competing with
transit traffic where queuing and data plane were not properly designed to
recognize control plane packets and handle it differently then transit the
moment packet hits the electrons on ingress int.

Thx
r.

Reply via email to