Re: Searching for the cause of a bad plan

From: Csaba Nagy <nagy(at)ecircle-ag(dot)com>
To: Ron Mayer <rm_pg(at)cheapcomplexdevices(dot)com>
Cc: postgres performance list <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Searching for the cause of a bad plan
Date: 2007-09-28 08:12:10
Message-ID: 1190967130.5430.85.camel@PCD12478
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Thu, 2007-09-27 at 11:07 -0700, Ron Mayer wrote:
> Csaba Nagy wrote:
> >
> > Well, my problem was actually solved by rising the statistics target,
>
> Would it do more benefit than harm if postgres increased the
> default_statistics_target?
>
> I see a fair number of people (myself included) asking questions who's
> resolution was to ALTER TABLE SET STATISTICS; and I think relatively
> fewer (if any?) people concerned about space in pg_statistic
> or people improving analyze time by reducing the statistics target.

Well, the cost of raising the statistics target is far from zero: with
all defaults the analyze time was ~ 10 seconds, with one column set to
100 was ~ 1.5 minutes, with one column set to 1000 was 15 minutes for
the table in question (few 100M rows). Of course the IO load must have
been proportional to the timings... so I'm pretty sure the current
default is serving well most of the situations.

Cheers,
Csaba.

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Csaba Nagy 2007-09-28 12:33:52 Re: Searching for the cause of a bad plan
Previous Message Richard Huxton 2007-09-28 08:06:11 Re: sequence query performance issues