From: Josh Berkus <josh@agliodbs.com>
To: pgsql-performance@postgresql.org
Cc: Brendan Duddridge <brendan@clickspace.com>
Subject: Re: Query using SeqScan instead of IndexScan
Date: Wed, 29 Mar 2006 21:18:49 -0800
Message-ID: <200603292118.50033.josh@agliodbs.com> (raw)
In-Reply-To: <2F1712F1-E5E2-455E-A291-06F300A66549@clickspace.com>
References: <2F1712F1-E5E2-455E-A291-06F300A66549@clickspace.com>
Brenden,
> Any ideas what I can do to improve this without turning sequential
> scanning off?
Hmmm, looks like your row estimates are good. Which means it's probably your
postgresql.conf parameters which are off. Try the following, in the order
below:
1) Raise effective_cache_size to 2/3 of your RAM (remember that ecs is in 8k
pages). Test again.
2) Multiply all of the cpu_* costs by 0.3. Test again.
3) Lower random_page_cost by steps to 3.5, then 3.0, then 2.5, then 2.0,
testing each time.
These are all runtime-settable parameters, so you can test them in one query
window, then set them in the main postgresql.conf if they work.
--
Josh Berkus
Sun Microsystems
San Francisco
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: josh@agliodbs.com, brendan@clickspace.com
Subject: Re: Query using SeqScan instead of IndexScan
In-Reply-To: <200603292118.50033.josh@agliodbs.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox