pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: csajl <csajl@yahoo.com>
To: josh@agliodbs.com
To: pgsql-performance@postgresql.org
Subject: Re: [repost] partial index / funxtional idx or bad sql?
Date: Mon, 12 May 2003 17:47:10 -0700 (PDT)
Message-ID: <20030513004710.58428.qmail@web40312.mail.yahoo.com> (raw)
In-Reply-To: <200305121713.38449.josh@agliodbs.com>


hi josh.

i'm using 7.3.2.  i tried using EXISTS instead of the IN, but the same query
now returns in seven sceonds as opposed to four with the IN.


cmdb=# EXPLAIN ANALYZE
cmdb-# select c.class_id, c.areacode, c.title from classifieds c
cmdb-# where c.class_cat_id = '1'
cmdb-# and c.areacode IN (
cmdb(# select areacode from cm_areacode where site_id = '10')
cmdb-# ;
                                                                    QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------
 Index Scan using classifieds_dual_idx on classifieds c  (cost=0.00..26622.14
rows=1837 width=39) (actual time=345.48..2305.04 rows=8460 loops=1)
   Index Cond: (class_cat_id = 1)
   Filter: (subplan)
   SubPlan
     ->  Materialize  (cost=3.46..3.46 rows=4 width=2) (actual time=0.00..0.01
rows=5 loops=61966)
           ->  Index Scan using site_cm_areacode_idx on cm_areacode 
(cost=0.00..3.46 rows=4 width=2) (actual time=0.14..0.22 rows=5 loops=1)
                 Index Cond: (site_id = 10)
 Total runtime: 2314.14 msec
(8 rows)
----------------------------------

classifieds_dual_idx  is the btree index on (class_type_id, areacode)
and site_cm_areacode_idx is the btree index on (site_id) only.
there is an index in the areacode table that has both (site_id, areacode) but
it's apparently not being used.  would it help the query to use that index
instead?

thanks for your help.



--- Josh Berkus <josh@agliodbs.com> wrote:
> Seth,
> 
> > SELECT p.id, p.areacode, p.content 
> > FROM posts p
> > WHERE p.type_id = ?
> > AND p.areacode in (
> >   select areacode from areacodes
> >    where site_id = ?
> >  )
> 
> Unless you're using 7.4 from CVS, you want to get rid of that IN:
> 
>  SELECT p.id, p.areacode, p.content 
>  FROM posts p
>  WHERE p.type_id = ?
>  AND EXISTS (
>   select areacode from areacodes
>     where site_id = ?
>     and p.areacode = areacodes.areacode
>   );
> 
> See how that works, and if it's still slow, post the EXPLAIN ANALYZE.


__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com




view thread (10+ messages)  latest in thread

Message-ID: <20030513004710.58428.qmail@web40312.mail.yahoo.com>
Permalink:  ../20030513004710.58428.qmail@web40312.mail.yahoo.com/
Also on:    postgresql.org/message-id/20030513004710.58428.qmail@web40312.mail.yahoo.com

 · 

reply

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 from your mail client; the link fills in
  the headers below: reply via email
  Some clients, webmail especially, drop In-Reply-To and References from
  a link, and the reply then starts a new thread: use the mbox below.

  To: pgsql-performance@postgresql.org
  Cc: csajl@yahoo.com, josh@agliodbs.com
  Subject: Re: [repost] partial index / funxtional idx or bad sql?
  In-Reply-To: <20030513004710.58428.qmail@web40312.mail.yahoo.com>
  References: <200305121713.38449.josh@agliodbs.com> <20030513004710.58428.qmail@web40312.mail.yahoo.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