Re: anti-join chosen even when slower than old plan

From: bricklen <bricklen(at)gmail(dot)com>
To: Marc Mamin <M(dot)Mamin(at)intershop(dot)de>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: anti-join chosen even when slower than old plan
Date: 2010-11-13 16:15:06
Message-ID: AANLkTimLhP0Ja_gETnSoW9uuQGS0ETGFZMztpxeaAhkh@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Sat, Nov 13, 2010 at 1:32 AM, Marc Mamin <M(dot)Mamin(at)intershop(dot)de> wrote:
> Hello,
>
> Just a short though:
>
> Is it imaginable to compare the prognoses of the plans with the actual
> results
> and somehow log the worst cases ?
>
> a) to help the DBA locate bad statistics and queries
> b) as additional information source for the planner
>
> This could possibly affect parameters of your formula on the fly.
>
> best regards,
>
> Marc Mamin

The contrib module auto_explain might help out here if you wanted to
roll your own solution for plan comparison.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2010-11-13 16:42:00 Re: temporary tables, indexes, and query plans
Previous Message Jon Nelson 2010-11-13 16:14:54 Re: temporary tables, indexes, and query plans