Date: Wed, 30 May 2007 13:49:43 -0700 From: security <security@jim-liesl.org> To: Andrew Gallatin <gallatin@cs.duke.edu> Cc: freebsd-performance@freebsd.org Subject: Re: [Fwd: asymetric speeds over gigE link] Message-ID: <465DE367.8070009@jim-liesl.org> In-Reply-To: <18013.49538.356407.283631@grasshopper.cs.duke.edu> References: <4652383E.9000302@jim-liesl.org> <18013.49538.356407.283631@grasshopper.cs.duke.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> security writes:
> > Sent this to -net and didn't get much info, so I'll try here since
> > there's some overlap
> >
> > Summary: Using iperf to measure TCP net speed between a linux (kubuntu edgy) and
>
> Iperf is probably your problem, it tends to perform really poorly on
> nearly any OS other than linux as it measures lots of things besides
> network performance (like gettimeofday performance). Try a network
> benchmark that just benchmarks the network, like netperf.
>
> For example, on an amd64x2 running -current, I see ~4.75Gb/s from
> netperf (no options) and 2.85Gb/s from iperf (no options)
>
> Drew
>
Drew,
Odd you should mention netperf. It yielded nearly identical
asymmetric results as iperf. I ended up testing with netpipe (NPtcp).
It reported fairly symmetric speeds between the two boxes when the send
and recv buffers were set the same, so that mystery is solved (maybe).
That being said I still wonder if I've too abstracted the net perfromance.
The down side is that the best speed I got was around 362Mbps on a
gigE link (with netpipe). I had hoped for better.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?465DE367.8070009>