>>> On Tue, Aug 21, 2007 at 1:12 PM, in message
<Pine.GSO.4.64.0708211316210.16511@westnet.com>, Greg Smith
<gsmith@gregsmith.com> wrote:
>
> The other thing I realized you haven't mentioned yet is what operating
> system you're using.
Linux version 2.6.5-7.286-bigsmp (geeko@buildhost) (gcc version 3.3.3 (SuSE Linux)) #1 SMP Thu May 31 10:12:58 UTC 2007
SUSE LINUX Enterprise Server 9 (i586)
VERSION = 9
PATCHLEVEL = 3
I know this is old, but I was happy just to get them off of Windows. :-/
> My suggestion for a starting 8.3 configuration for you would be adjusting
> these settings as follows:
>
> shared_buffers=1GB
> checkpoint_segments = 50
>
> And then try replaying your data with checkpoint_completion_target at 0.5
> (default), 0.7, and 0.9 and see how each of those works out for you (0.7
> is the least useful of those if you only did two). Hopefully the data
> replay you have can be setup to invoke the same tightly spaced checkpoint
> behavior you commented about. Based on the results of those tests, it may
> be possible to further increase shared_buffers, and
> checkpoint_segments/checkpoint_timeout may need some adjustment one way or
> another.
>
> Note that some/all of the bgwriter_lru parameters may be going away before
> 8.3 is done as well, that's a loose end I'm working on right now.
Thanks much. If I get the approval, I'll be sure to run these three tests.
It probably makes sense to get a baseline with the current config first.
I assume you want me to do these with both the update and query streams
running full out?
Anyone else have a configuration to suggest?
-Kevin
