From: Josh Close <narshe@gmail.com>
To: POSTGRES-PERFORMANCE <pgsql-performance@postgresql.org>
Subject: Re: slow queries, possibly disk io
Date: Fri, 27 May 2005 10:00:23 -0500
Message-ID: <4a0cafe2050527080038616a2f@mail.gmail.com> (raw)
In-Reply-To: <14584.1117204158@sss.pgh.pa.us>
References: <4a0cafe205052613242e0d4d10@mail.gmail.com>
<758d5e7f05052614232f6c9164@mail.gmail.com>
<4a0cafe205052706043ef8f53f@mail.gmail.com>
<4a0cafe205052706042d4e60e5@mail.gmail.com>
<14584.1117204158@sss.pgh.pa.us>
Doing the query
explain
SELECT COALESCE( SUM( iNumSent ), 0 ) AS iNumSent
FROM adaption.tblBatchHistory_1
WHERE tStamp > ( now() - interval '5 mins' )::text
gives me this:
Aggregate (cost=32138.33..32138.33 rows=1 width=4)
-> Seq Scan on tblbatchhistory_1 (cost=0.00..31996.10 rows=56891 width=4)
Filter: ((tstamp)::text > ((now() - '00:05:00'::interval))::text)
Still not an index scan.
On 5/27/05, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Josh Close <narshe@gmail.com> writes:
> > this_sQuery := \'
> > SELECT COALESCE( SUM( iNumSent ), 0 ) AS iNumSent
> > FROM adaption.tblBatchHistory_\' || this_rServerIds.iId || \'
> > WHERE tStamp > now() - interval \'\'5 mins\'\';
> > \';
>
> > Here is the explain analyze of one loops of the sum:
>
> > Aggregate (cost=31038.04..31038.04 rows=1 width=4) (actual
> > time=14649.602..14649.604 rows=1 loops=1)
> > -> Seq Scan on tblbatchhistory_1 (cost=0.00..30907.03 rows=52401
> > width=4) (actual time=6339.223..14648.433 rows=919 loops=1)
> > Filter: (tstamp > (now() - '00:05:00'::interval))
> > Total runtime: 14649.709 ms
>
> I think you really want that seqscan to be an indexscan, instead.
> I'm betting this is PG 7.4.something? If so, probably the only
> way to make it happen is to simplify the now() expression to a constant:
>
> SELECT COALESCE( SUM( iNumSent ), 0 ) AS iNumSent
> FROM adaption.tblBatchHistory_\' || this_rServerIds.iId || \'
> WHERE tStamp > \\\'' || (now() - interval \'5 mins\')::text ||
> \'\\\'\';
>
> because pre-8.0 the planner won't realize that the inequality is
> selective enough to favor an indexscan, unless it's comparing to
> a simple constant.
>
> (BTW, 8.0's dollar quoting makes this sort of thing a lot less painful)
>
> regards, tom lane
>
--
-Josh
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: narshe@gmail.com
Subject: Re: slow queries, possibly disk io
In-Reply-To: <4a0cafe2050527080038616a2f@mail.gmail.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