Content-Type: text/plain; charset="utf-8"
Date: Tue, 04 Dec 2012 14:09:20 -0500
From: "Kevin Grittner" <kgrittn@mail.com>
Message-ID: <20121204190920.142860@gmx.com>
MIME-Version: 1.0
Subject: Re: Slow query: bitmap scan troubles
To: postgresql@foo.me.uk,"postgres performance list"
	<pgsql-performance@postgresql.org>
Content-Transfer-Encoding: 8bit
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

postgresql@foo.me.uk wrote:

> Ah okay, thanks. I knew I could set various things but not
> effective_work_mem (I tried reloading the edited config file but
> it didn't seem to pick it up)

Check the server log, maybe there was a typo or capitalization
error.

To test on a single connection you should be able to just run:

SET effective_cache_size = '88GB';

By the way, one other setting that I have found a need to adjust to
get good plans is cpu_tuple_cost. In my experience, better plans
are chosen when this is in the 0.03 to 0.05 range than with the
default of 0.01.

-Kevin


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance