[httperf] file descriptors
mnot at yahoo-inc.com
Tue Jul 29 16:34:10 PDT 2008
I've compared against the select-based httperf against a few known
servers, and the results seem equivalent.
The interesting thing is that at lower rates, CPU usage is high (but
not >90%, as with select), but as they get higher, it dips pretty
substantially (much < 50%) and then slowly ramps up again. This is
useful, because you can tell when CPU on the client may become an
I think we should get this out there ASAP. The only thing that needs
to happen now is that configure should complain if it can't find
libevent, I think, and the docs need to be updatd WRT CPU usage.
The biggest perf limit I've got now is that I can generate about 550
*connections* a second; past that gets me a connection failed with
unexpected error 99, which is ENOSTR (not a stream). This is Red Hat
Linux. Any thoughts here? I see this with both select and libevent, so
it's not the new code. It always happens around the same rate; it's
not dependent upon how long the test has run, or the timeout
(suggesting that it's not cumulative resource exhaustion a la
On 09/07/2008, at 8:42 AM, Adrian Chadd wrote:
> On Wed, Jul 09, 2008, Theodore Bullock wrote:
>>> Adrian Chadd wrote:
>>> (Noone's bothered trying it out? Its going to be a hell of a lot
>>> than stock httperf..)
>> I will try it out this weekend.
>> I have been working on a virtual machine with 256 mb of memory, so
>> things area ridiculously slow until I get my laptop repaired.
> Ok. Please let me know. I'd ideally like my work to be folded into
> the main httperf distribution.
> httperf mailing list
> httperf at linux.hpl.hp.com
Mark Nottingham mnot at yahoo-inc.com
More information about the httperf