Re: Shouldn't the planner have a higher cost for reverse index scans?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Merlin Moncure <mmoncure(at)gmail(dot)com>
Cc: Lists <lists(at)on-track(dot)ca>, Josh Berkus <josh(at)agliodbs(dot)com>, postgres performance list <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Shouldn't the planner have a higher cost for reverse index scans?
Date: 2009-04-16 15:36:34
Message-ID: 18553.1239896194@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Merlin Moncure <mmoncure(at)gmail(dot)com> writes:
> On Thu, Apr 16, 2009 at 2:02 AM, Lists <lists(at)on-track(dot)ca> wrote:
>> select comment_date
>> from user_comments
>> where user_comments.uid=1
>> order by comment_date desc limit 1

> try this:
> create index comment_data_uid_idx on user_comments(uid, comment_date);

> select * from user_comments where (uid, comment_date) < (1, high_date)
> order by uid desc, comment_date desc limit 1;

You don't really need to complicate your queries like that. Having the
two-column index will suffice to make the given query work fine, at
least in reasonably modern PG versions (since 8.1 I think).

regards, tom lane

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2009-04-16 15:42:43 Re: Shouldn't the planner have a higher cost for reverse index scans?
Previous Message Merlin Moncure 2009-04-16 14:44:03 Re: Really dumb planner decision