Optimizing for high PPS, Intel NICs

[Ivan Voras]

Ivan Voras wrote:

> I've noticed something strange: the server is bottlenecked with "em1
> taskq" kernel thread taking 100% of a CPU core, while the global CPU
> utilization is around 50%, but the client's em0 taskq thread for this
> same load is ~~ 10% (with > 30% idle). The client CPU is a bit faster
> then the server (2.4 GHz vs 2.0 GHz) but I don't think this can account
> for such a big difference. Toggling TSO on the server doesn't help.

I've switched the server and the client role and the behaviour is always
the same - on this one machine the taskq starts using 100% of a core
when pushing more than about 150,000 PPS. It's the same when testing
under Linux so it looks like I need to shop for a better NIC.

Can anyone recommend a good but basic (no fancy features needed) PCI-E
or PCI-X NIC that's known to be able to push > 500,000 PPS?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 258 bytes
Desc: OpenPGP digital signature
Url : http://lists.freebsd.org/pipermail/freebsd-net/attachments/20080930/bf8a3a21/signature.pgp