Re: Performance degradation of inserts when database size grows

On 05/17/2011 07:45 AM, Andrey Vorobiev wrote:

> 2011-05-17 18:55:51 NOVST LOG: checkpoint starting: xlog
> 2011-05-17 18:57:20 NOVST LOG: checkpoint complete: wrote 62861 buffers
> (24.0%); 0 transaction log file(s) added, 0 removed, 0 recycled;
> write=89.196 s, sync=0.029 s, total=89.242 s

Increase your checkpoint_segments. If you see "checkpoint starting:
xlog" instead of "checkpoint starting: time", you don't have enough
checkpoint segments to handle your writes. Checkpoints *will* degrade
your throughput.

--
Shaun Thomas
OptionsHouse | 141 W. Jackson Blvd. | Suite 800 | Chicago IL, 60604
312-676-8870
sthomas@peak6.com

______________________________________________

See  http://www.peak6.com/email_disclaimer.php
for terms and conditions related to this email