pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Contact AR-SD.NET <contact@ar-sd.net>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-performance@postgresql.org
Subject: Re: Index not used in query. Why?
Date: Tue, 19 Oct 2004 20:49:45 +0300
Message-ID: <0c4501c4b604$06f15460$0b00a8c0@forge> (raw)
References: <0bef01c4b5f8$70026ae0$0b00a8c0@forge>
	<16865.1098204769@sss.pgh.pa.us>

Is there a solution to make it faster?
At the end I need only in the query the id_status =4 and 6, but if I write
in the sql query (where condition) where id_status in (4,6), the explain
says the same(the slow version).

For example:
SELECT count(o.id)  FROM orders o
                                   INNER JOIN report r ON o.id=r.id_order
                                   INNER JOIN status s ON o.id_status=s.id
                                   INNER JOIN contact c ON o.id_ag=c.id
                                   INNER JOIN endkunde e ON
o.id_endkunde=e.id
                                   INNER JOIN zufriden z ON
r.id_zufriden=z.id
                                   INNER JOIN plannung v ON
v.id=o.id_plannung
                                   INNER JOIN mpsworker w ON
v.id_worker=w.id
                                   INNER JOIN person p ON p.id = w.id_person
                                   WHERE o.id_status in (4,6);

The result for this query is also without index searches.

I really have to make this query a little more faster. Suggestions?

Regards,
Andy.

----- Original Message -----
From: "Tom Lane" <tgl@sss.pgh.pa.us>
To: "Andrei Bintintan" <klodoma@ar-sd.net>
Cc: <pgsql-performance@postgresql.org>
Sent: Tuesday, October 19, 2004 7:52 PM
Subject: Re: [PERFORM] Index not used in query. Why?


> "Andrei Bintintan" <klodoma@ar-sd.net> writes:
> > Hi to all! I have the following query. The execution time is very big,
it
> > doesn't use the indexes and I don't understand why...
>
> Indexes are not necessarily the best way to do a large join.
>
> > If I use the following query the indexes are used:
>
> The key reason this wins seems to be that the id_status = 4 condition
> is far more selective than id_status > 3 (the estimates are 52 and 36967
> rows respectively ... is that accurate?) which means that the second
> query is inherently about 1/700th as much work.  This, and not the use
> of indexes, is the fundamental reason why it's faster.
>
> regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster
>




view thread (4+ messages)  latest in thread

Message-ID: <0c4501c4b604$06f15460$0b00a8c0@forge>
Permalink:  ../0c4501c4b604$06f15460$0b00a8c0@forge/
Also on:    postgresql.org/message-id/0c4501c4b604$06f15460$0b00a8c0@forge

 · 

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 using the --to and --cc options:
  reply via email

  To: pgsql-performance@postgresql.org
  Cc: contact@ar-sd.net, tgl@sss.pgh.pa.us
  Subject: Re: Index not used in query. Why?
  In-Reply-To: <0c4501c4b604$06f15460$0b00a8c0@forge>

* 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