Kevin Grittner wrote:
> I've been lobbying management here for us to allocate some resources to testing 8.3 once it hits beta. If it is
approved,it might happen on a time frame too short to get much feedback before the tests, so I'm throwing the question
outhere now: what would people like us to bang on?
That would be great!
> We're particularly interested in seeing what configuration changes we may have to make to achieve optimal performance
withthe checkpoints and background writer in the new release. When we first went to PostgreSQL our biggest problem was
thatdirty buffers would accumulate in shared memory until a checkpoint, and then overrun the controllers cache. This
wouldcause disk reads to queue up behind the writes, and queries which normally ran in a millisecond or two were timing
outat our renderers' 20 second limit. The problem went away completely when we used a very aggressive background
writerconfiguration, to put the dirty pages in front of the OS file system right away, so that its algorithms and the
controllercache could deal with things before they got out of hand.
Yes, the load distributed checkpoints definitely should help with that.
I'd like to see how well it works for you with the default bgwriter
settings.
-- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com