[lkml]   [2008]   [Jul]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Higher than expected disk write(2) latency

> I thought you were doing I/O to the underlying block device. If so,
> there's no need to open with O_SYNC. You do, however, need to open the
> device with O_DIRECT and align your buffers (and buffer lengths)
> properly.

Yes, we are using O_DIRECT + aligning the buffers. However, usign O_SYNC
vs. O_ASYNC seems to matter (different latency). Maybe this is because
we are using librt?

> Which AIO interface are you using, libaio or librt? How many I/Os are
> you queueing to the device? You may want to take a look at aio-stress.c
> as a way to test your device (this uses libaio, the in-kernel AIO
> interface).

We are using librt, queueing 10240 writes, 512 bytes each.

We'll retry with libaio. Thanks for advice.


 \ /
  Last update: 2008-07-02 20:23    [W:0.095 / U:0.412 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site