* Suggestion on size= and test length
@ 2011-01-02 4:12 Spelic
2011-01-03 11:04 ` Jens Axboe
0 siblings, 1 reply; 2+ messages in thread
From: Spelic @ 2011-01-02 4:12 UTC (permalink / raw)
To: fio
Dear fio developers,
Thanks for the very good application, but may I suggest that the size=
option controls only the size of the files used for the test (in
particular random r/w tests), while there should be another option
totaltransfer= which controls how long is the test? And/or a walltime=
option to make the test length time based... even better!
Currently we either are hit by short seeks (size= small) or have to wait
for impossibly long tests uselessly (size= large).
Thank you
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Suggestion on size= and test length
2011-01-02 4:12 Suggestion on size= and test length Spelic
@ 2011-01-03 11:04 ` Jens Axboe
0 siblings, 0 replies; 2+ messages in thread
From: Jens Axboe @ 2011-01-03 11:04 UTC (permalink / raw)
To: Spelic; +Cc: fio
On 2011-01-02 05:12, Spelic wrote:
> Dear fio developers,
>
> Thanks for the very good application, but may I suggest that the size=
> option controls only the size of the files used for the test (in
> particular random r/w tests), while there should be another option
> totaltransfer= which controls how long is the test? And/or a walltime=
> option to make the test length time based... even better!
>
> Currently we either are hit by short seeks (size= small) or have to
> wait for impossibly long tests uselessly (size= large).
What about just adding a runtime= setting to shorten the run? Use the
full size, but put a cap on runtime.
--
Jens Axboe
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2011-01-03 11:03 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-01-02 4:12 Suggestion on size= and test length Spelic
2011-01-03 11:04 ` Jens Axboe
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox