Re: Performance with ethernet channel bonding

From: Mark_H_Johnson@Raytheon.com
Date: Mon Aug 28 2000 - 09:22:24 EST


>Andi's message...
> There must be some misunderstanding. My point was that round robin
inherently
> leads to packet reordering (easy to see when the packets have different
lengths,
> but even with equal packet lengths small timing differences can cause
it).
> Packet reordering is bad for performance.

Ahh. Of course. I should have thought of that.

> The default 32K window is too small for fast networks (fast ethernet and
up,
> or anything with longer latency like round robin setups). You can change
it
> using the /proc/sys/net/core/[rw]mem_{default,max} sysctls (see
socket(7))

Hmm. The machine I'm running with has 65535 for [rw]mem_default &
[rw]mem_max. I assume this is a 32 bit value so I could go up to a megabyte
if needed to get the performance consistent on the bound channels. I also
assume I must make the same changes on all affected machines. I'll try some
more tests & get back with the results.

Thanks.
--Mark H Johnson
  <mailto:Mark_H_Johnson@raytheon.com>

-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org



This archive was generated by hypermail 2b29 : Thu Aug 31 2000 - 21:00:31 EST