[nsp] Throughput of 100 Mbps

[Terry Baranski]

> You are raising a very interesting question. I never understood
> (or seriously tried to understand) what is meant by this
> throughput numbers.
> 
> Are you saying that the pps numbers from Cisco are always for 
> 64-bytes packet? Could be, in fact this gives the highest pps
> number... And it's always one-way, right?

Cisco's numbers are all for 64-byte packets to my knowledge, as are most
pps numbers from other vendors.  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.

> The test I made was very simple: a 3620 with 2 FE interfaces, 
> and a PC on each of those; then FTP between the two PCs.
> 
> I only had 40% of CPU usage - why isn't it trying harder, to
> reach a higher throughput? :-) 

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. 

-Terry