Date: Wed, 10 May 2006 14:23:29 -0500
From: Bruno Wolff III <bruno@wolff.to>
To: ogjunk-pgjedan@yahoo.com
Cc: pgsql-sql@postgresql.org
Subject: Re: Help with a seq scan on multi-million row table
Message-ID: <20060510192329.GB4099@wolff.to>
Mail-Followup-To: Bruno Wolff III <bruno@wolff.to>,
	ogjunk-pgjedan@yahoo.com, pgsql-sql@postgresql.org
References: <20060510181359.58201.qmail@web50306.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20060510181359.58201.qmail@web50306.mail.yahoo.com>
User-Agent: Mutt/1.4.2.1i

On Wed, May 10, 2006 at 13:13:59 -0500,
  ogjunk-pgjedan@yahoo.com wrote:
>   Hello,
> 
> I have a little 2-table JOIN, GROUP BY, ORDER BY query that does a sequential scan on a multi-million row table.  I _thought_ I had all the appropriate indices, but apparently I do not.  I was wondering if anyone can spot a way I can speed up this query.
> The query currently takes... *gulp*: 381119.201 ms :(
>   
> There are only 2 tables in the game: user_url and user_url_tag.  The latter has FKs pointing to the former.  The sequential scan happens on the latter - user_url_tag:
> 
> EXPLAIN ANALYZE select DISTINCT userurltag0_.tag as x0_0_, COUNT(*) as x1_0_ from user_url_tag userurltag0_, user_url userurl1_ WHERE (((userurl1_.user_id=1 )) AND ((userurltag0_.user_url_id=userurl1_.id ))) GROUP BY userurltag0_.tag ORDER BY  count(*) DESC;  

While this isn't a big issue, it looks like DISTINCT is redundant in your
query and seems to be adding some extra work.