From: | "Michael Paesold" <mpaesold(at)gmx(dot)at> |
---|---|
To: | "Mark Wong" <markw(at)osdl(dot)org>, <pgsql-hackers(at)postgresql(dot)org>, <testperf-general(at)pgfoundry(dot)org> |
Subject: | Re: postgresql CVS callgraph data from dbt2 |
Date: | 2005-09-14 08:01:30 |
Message-ID: | 01a201c5b902$aa76c230$0f01a8c0@zaphod |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Mark Wong wrote:
> Hi everyone,
>
> For those of you watching the the daily results generated from STP
> (http://developer.osdl.org/markw/postgrescvs/dbt2/) I have callgraph
> data from oprofile collected starting from the Sept 9 results. Here is
> an example of what it looks like:
>
> http://www.testing.osdl.org/stp/303184/results/0/callgraph.txt
>
> I hope it's useful.
Rather unrelated, but the tests here use the standard settings for
wal_buffers. As I have read on this list, 8.1 needs a much higher setting
for wal_buffers for good performance (128 or 256?). I don't know if you want
to change this setting because it will make the results incomparable, but it
should improve overall performance.
See
http://archives.postgresql.org/pgsql-hackers/2005-07/msg01004.php
Best Regards,
Michael Paesold
From | Date | Subject | |
---|---|---|---|
Next Message | Andreas Pflug | 2005-09-14 08:18:57 | Re: 8.1 system info / admin functions |
Previous Message | Michael Paesold | 2005-09-14 07:41:46 | Re: Spinlocks, yet again: analysis and proposed patches |