From: Tom Lane <tgl@sss.pgh.pa.us>
To: PFC <lists@boutiquenumerique.com>
Cc: Ken Egervari <ken@upfactor.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Performance problem with semi-large tables
Date: Sun, 30 Jan 2005 12:50:02 -0500
Message-ID: <3657.1107107402@sss.pgh.pa.us> (raw)
In-Reply-To: <opslexchx2th1vuj@musicbox>
References: <001601c50416$57090e10$cd422418@a96dfxb4kjzogw>
<opsldw48czth1vuj@musicbox>
<002201c50654$1a8c50b0$cd422418@a96dfxb4kjzogw>
<opsld0ul0nth1vuj@musicbox>
<000a01c5066a$0d1f43e0$cd422418@a96dfxb4kjzogw>
<opslexchx2th1vuj@musicbox>
PFC <lists@boutiquenumerique.com> writes:
>> For example, let's add another filter to get all the shipments with
>> release code '5' that are 7 days old or newer.
>>
>> ss.date >= current_date - 7
> It's the order by + limit which makes the query behaves badly, and which
> forces use of kludges to use the index. If you add another condition like
> that, it should be a breeze.
Actually, that date condition has its own problem, namely that the
compared-to value isn't a constant. The 8.0 planner is able to realize
that this is a pretty selective condition, but prior releases fall back
on a very pessimistic default estimate. I'm sure that has something to
do with Ken not being able to get it to use an index on date.
regards, tom lane
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, lists@boutiquenumerique.com, ken@upfactor.com
Subject: Re: Performance problem with semi-large tables
In-Reply-To: <3657.1107107402@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