Message-ID: <4DB88690.5050701@selectacast.net>
Date: Wed, 27 Apr 2011 17:11:44 -0400
From: Joseph Shraibman <jks@selectacast.net>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
	rv:1.9.2.15) Gecko/20110307 Fedora/3.1.9-0.38.b3pre.fc13
	Thunderbird/3.1.9
MIME-Version: 1.0
To: Robert Haas <robertmhaas@gmail.com>
CC: pgsql-performance@postgresql.org
Subject: Re: index usage on queries on inherited tables
References: <in36ut$24hs$1@news.hub.org>
	<BANLkTi=g=dA7Bt_P=m0Amknp9r-k6x5-Sw@mail.gmail.com>
In-Reply-To: <BANLkTi=g=dA7Bt_P=m0Amknp9r-k6x5-Sw@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Received-SPF: Pass (client IP white listed); receiver=corp1.xtenit.com;
	client-ip=66.166.198.124; envelope-from=<jks@selectacast.net>
Received-SPF: Pass (client IP white listed); receiver=corp1.xtenit.com;
	client-ip=66.166.198.124; helo=
Received-SPF: Pass (sender authenticated); receiver=tupari.net;
	client-ip=192.168.1.3; envelope-from=<jks@selectacast.net>
Received-SPF: Pass (sender authenticated); receiver=tupari.net;
	client-ip=192.168.1.3; helo=

On 04/27/2011 04:32 PM, Robert Haas wrote:
> In the first case, PostgreSQL evidently thinks that using the indexes
> will be slower than just ignoring them.  You could find out whether
> it's right by trying it with enable_seqscan=off.

My point is that this is just a problem with inherited tables.  It
should be obvious to postgres that few rows are being returned, but in
the inherited tables case it doesn't use indexes.  This was just an
example.  In a 52 gig table I have a "select id from table limit 1 order
by id desc" returns instantly, but as soon as you declare a child table
it tries to seq scan all the tables.