Hello,

On Wed, Mar 25, 2009 at 4:10 AM, Rocco Caputo <[email protected]> wrote:
>
> time(1) includes Perl & Python startup/shutdown time, which may be
> significant.
>

Suppose the difference is negligible.


> Your times seem to be wall clock based, which can vary based on system load,
> network conditions, and so on.  Show the CPU times instead.
>
> It appears that you've only run one iteration.  Benchmarks should be run
> enough times to account for statistical variance.
>

No, I run the test for several times, on an idle server (Intel Quad
Core, SMP, 4GB RAM. using both latest POE & Twisted). The difference
is roughly 200% differences, so I was surprised.


Thank you.

Reply via email to