[netperf-talk] Netperf & iPerf

[Tom Wambold]

On 1/22/07, Rick Jones <rick.jones2 at hp.com> wrote:
> Tom Wambold wrote:
> > Dear all,
> >
> > Can anyone explain the differences between Netperf and iPerf
> > throughput tests?  I recently did some throughput tests with some
> > 802.11b cards and laptops, and found that Netperf was reporting on
> > average 1.5Mbs higher throughput than iPerf.
> >
> > If anyone can give some insight on this, I'd appreciate it.
>
> What sort of command lines were used for each test?
>
> By default, netperf passes what could be very compressible buffers to
> the stack - dies 802.11b try to do any compression?
>
> BY default netperf accepts the system's default socket buffer and thus
> window size.  Does iperf do that, or have its own default?
>
> soooo many possibilities, we need to know more about your tests to take
> more accurate guesses.
>
> happy benchmarking,
>
> rick jones
>

We ran 2 tests, TCP and UDP stream tests.  The commands used were:
netperf -H <hostip> -t UDP_STREAM -- -m 1024
netperf -H <hostip> -t TCP_STREAM -- -m 1024

802.11b does not do any sort of compression to the best of my knowledge.

iPerf seems to only factor in data received at the application layer.
Does Netperf throughput calculations include all data sent (network
headers), or just what is received at the application layer?

What other information could you use?