(2013/10/08 17:33), Haribabu kommi wrote:
> The checkpoint_timeout and checkpoint_segments are increased to make sure no checkpoint happens during the test run.
Your setting is easy occurred checkpoint in checkpoint_segments = 256. I don't
know number of disks in your test server, in my test server which has 4 magnetic
disk(1.5k rpm), postgres generates 50 - 100 WALs per minutes.
And I cannot understand your setting which is sync_commit = off. This setting
tend to cause cpu bottle-neck and data-loss. It is not general in database usage.
Therefore, your test is not fair comparison for Fujii's patch.
Going back to my DBT-2 benchmark, I have not got good performance (almost same
performance). So I am checking hunk, my setting, or something wrong in Fujii's
patch now. I am going to try to send test result tonight.
Regards,
--
Mitsumasa KONDO
NTT Open Source Software Center