[Xorp-users] Why Xorp so slow ?

[Sébastien Namèche]

Bruce,

Thanks for this fast answer.


Le 3 juil. 09 à 13:51, Bruce Simpson a écrit :
> What exactly is the specification of the hardware you're using, to  
> give us a better feel for the slowdown? Did you perform these  
> comparisons on the same hardware?

Yes, same hardware and same OS. We did it for FreeBSD and Linux (it  
seems an little faster under Linux).

The hardware is an AMD Athlon XP at 1500 MHz (bus clock is 100 MHz)  
with 1,5 Gb RAM.
For production, we plan to use a Pentium 4 at 2800 MHz (bus clock 800  
MHz) with 1 Gb RAM.


> The danger in informal benchmarks is that it's easy to trip up over  
> things like the L2 cache being primed, kernel memory fragmentation,  
> etc.

Yes, this is not a real, full, formal benchmark. This is rather  
"informal". But the difference is huge.


> Do you plan to deploy XORP? We can get performance up, however, it's  
> also a community effort at the end of the day.

Yes. We want to deploy it. I like the way it is possible to filter in  
Xorp.
We were able to dot things that were not possible with Quagga and  
OpenBGP/OpenOSPF.
(The only thing we couldn't do is the AS prepending for OSPF injected  
procefixes.)

OpenBGP/OpenOSPF are very fast and easy to configure, but have poor  
functionnalities.
But OpenOSPF is young...


Regards,

-- 
Seb

“Un peuple prêt à sacrifier un peu de liberté pour un peu de sécurité
ne mérite ni l’une ni l’autre, et finit par perdre les deux.”
-- Thomas Jefferson.