From: Steinar H. Gunderson <sgunderson@bigfoot.com>
To: Guillaume Smet <guillaume.smet@openwide.fr>
Cc: pgsql-performance@postgresql.org
Subject: Re: Problem analyzing explain analyze output
Date: Mon, 24 Oct 2005 02:08:24 +0200
Message-ID: <20051024000824.GA11800@samfundet.no> (raw)
In-Reply-To: <435C2297.5050207@openwide.fr>
References: <435C2297.5050207@openwide.fr>
On Mon, Oct 24, 2005 at 01:53:59AM +0200, Guillaume Smet wrote:
> I don't understand why I have the Nested Loop at line 19 with an actual
> time of 254.292..257.328 because I can't find anywhere the line taking
> this 254 ms.
You don't have a nested loop with that time; however, you have
> -> Nested Loop (cost=887.45..4031.09 rows=587 width=20) (actual time=254.424..280.794 rows=514 loops=1)
> -> Bitmap Heap Scan on contcrilieu ccl (cost=887.45..1668.96 rows=587 width=8) (actual time=254.292..257.328 rows=514 loops=1)
> Recheck Cond: ((dcrilieu >= (now() - '60 days'::interval)) AND ((flagcriaccepteelieu)::text = 'O'::text))
> -> Bitmap Index Scan on idx_contcrilieu_4 (cost=0.00..887.45 rows=587 width=0) (actual time=254.143..254.143 rows=514 loops=1)
> Index Cond: ((dcrilieu >= (now() - '60 days'::interval)) AND ((flagcriaccepteelieu)::text = 'O'::text))
> -> Index Scan using pk_lieu on lieu l (cost=0.00..4.01 rows=1 width=12) (actual time=0.034..0.036 rows=1 loops=514)
> Index Cond: ("outer".numlieu = l.numlieu)
which seems to make sense; you have one run of about 257ms, plus 514 runs
taking about 0.035ms each (ie. about 18ms), which should add up to become
about 275ms (which is close enough to the reality of 281ms).
/* Steinar */
--
Homepage: http://www.sesse.net/
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: sgunderson@bigfoot.com, guillaume.smet@openwide.fr
Subject: Re: Problem analyzing explain analyze output
In-Reply-To: <20051024000824.GA11800@samfundet.no>
* 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