pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Tom Lane <tgl@sss.pgh.pa.us>
To: sarlav kumar <sarlavk@yahoo.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: help needed -- sequential scan problem
Date: Sat, 20 Nov 2004 00:19:46 -0500
Message-ID: <16679.1100927986@sss.pgh.pa.us> (raw)
In-Reply-To: <20041119223122.49470.qmail@web51301.mail.yahoo.com>
References: <20041119223122.49470.qmail@web51301.mail.yahoo.com>

sarlav kumar <sarlavk@yahoo.com> writes:
> I have a query which does not use index scan unless I force postgres to use index scan. I dont want to force postgres, unless there is no way of optimizing this query.

The major issue seems to be in the sub-selects:

>                  ->  Seq Scan on merchant_purchase mp  (cost=0.00..95.39 rows=44 width=4) (actual time=2.37..2.58 rows=6 loops=619)
>                        Filter: (merchant_id = $0)

where the estimated row count is a factor of 7 too high.  If the
estimated row count were even a little lower, it'd probably have gone
for an indexscan.  You might get some results from increasing the
statistics target for merchant_purchase.merchant_id.  If that doesn't
help, I'd think about reducing random_page_cost a little bit.

			regards, tom lane



view thread (6+ messages)  latest in thread

Message-ID: <16679.1100927986@sss.pgh.pa.us>
Permalink:  ../16679.1100927986@sss.pgh.pa.us/
Also on:    postgresql.org/message-id/16679.1100927986@sss.pgh.pa.us

 · 

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: tgl@sss.pgh.pa.us, sarlavk@yahoo.com
  Subject: Re: help needed -- sequential scan problem
  In-Reply-To: <16679.1100927986@sss.pgh.pa.us>

* 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