Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Oct 2007 16:12:00 +0200
From:      "Petr Holub" <hopet@ics.muni.cz>
To:        "Adrian Chadd" <adrian@freebsd.org>
Cc:        performance@freebsd.org
Subject:   RE: Myrinet 10GE performance on 7.0-CURRENT
Message-ID:  <041b01c80cd9$db2b3e60$5317fb93@KLOBOUCEK>
In-Reply-To: <d763ac660710120601p1f2e6fbau569ec62dbc182eea@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
> Thanks for that. So its showing ~ 600k pps throughput. I'd love to
> know whats setting that upper limit - is it iperf? Userspace?
> Interrupt thread/device driver? Other kernel stuff?

When measuring using ipef, it is iperf itself with high probability,
as the netperf tool can achieve much better performance on small
packets. I plan to integrate iperf patch mentioned in Linux
kernel list to see if there's any difference.

Petr




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?041b01c80cd9$db2b3e60$5317fb93>