From: Tom Lane <tgl@sss.pgh.pa.us>
To: Abadie Lana <Lana.Abadie@iter.org>
cc: "pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Subject: Re: Very slow query (3-4mn) on a table with 25millions rows
In-reply-to: <E544BB9A64ABD24DA201745FD316D94551122DDA@XCH2.iter.org>
References: <E544BB9A64ABD24DA201745FD316D94551122DDA@XCH2.iter.org>
Comments: In-reply-to Abadie Lana <Lana.Abadie@iter.org>
	message dated "Mon, 25 Jul 2016 15:26:46 -0000"
Date: Mon, 25 Jul 2016 14:06:57 -0400
Message-ID: <4248.1469470017@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

Abadie Lana <Lana.Abadie@iter.org> writes:
> I'm having a problem with a slow query - I tried several things to optimize the queries but didn't really help. The output of explain analyse shows sequential scan on a table of 25 million rows. Even though it is indexed and (I put a multi-column index on the fields used in the query), the explain utility shows no usage of the scan...

That index looks pretty useless judging from the rowcounts, so I'm not
surprised that the planner didn't use it.  You might have better luck with
an index on the split_part expression

split_part(split_part((s.attvalue)::text, ' '::text, 1), '.'::text, 1)

since it's the join of that to e.name that seems to be actually selective.
(The planner doesn't appear to realize that it is, but ANALYZE'ing after
creating the index should fix that.)

			regards, tom lane


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance