[nsp] Throughput of 100 Mbps

Vincent De Keyzer vincent at dekeyzer.net
Mon Nov 10 10:33:21 EST 2003

> > And it's always one-way, right?
> I don't know
> whether or not the tests are run with a unidirectional flow, 
> but it may not matter from the router's perspective -- the 
> router is just forwarding packets, regardless of direction.

What I (quite obscurely I agree) meant is the following: 
I've heard that when some people see a router forwarding 20 kpps between two
they count 20 kpps in + 20 kpps out = 40 kpps! But I imagine Cisco is more
honest than that...

> I suspect if you ran a dedicated bandwidth testing program
> like TTCP you'd get a higher number. (Applications aren't 
> always able to achieve 100% throughput on any given 
> end-to-end link.)  When saturated, my experience is that the 
> router's CPU will max out in interrupt context if CEF is enabled. 

True, with TTCP I could now reach 75 Mbps (down from 89 Mbps when the two
PCs are connected back-to-back), with CPU around 70%. This is not the 100
Mbps I had hoped for (sigh), but it is definitely a better measurement than
my previous 40 Mbps with FTP.

Thanks for the tip.


More information about the cisco-nsp mailing list