Re: Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)

From: Claudio Freire <klaussfreire(at)gmail(dot)com>
To: Eoghan Murray <eoghan(at)qatano(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)
Date: 2012-07-13 16:40:31
Message-ID: CAGTBQpb=Z2cj_v32bbNS45fxdL_qgMAzc0Tg2=Hhq8ha4E2kQw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Fri, Jul 13, 2012 at 1:28 PM, Eoghan Murray <eoghan(at)qatano(dot)com> wrote:
> Thank you Claudio,
>
> I haven't touched the 9.1 configuration (with the exception of toggling the
> enable_material setting). http://pastebin.com/nDjcYrUd
> As far as I can remember I haven't changed the 8.4 configuration:
> http://pastebin.com/w4XhDRX4

Maybe that's your problem. Postgres default configuration is not only
suboptimal, but also a poor reflection of your hardware (what's your
hardware, btw?). Which means postgres' expected costs won't hold. I'm
thinking especially about your effective_cache_size, which may
influence postgres' decision to use one join method vs another, but
many other settings would probable influence.

Spend a bit of time to configure both servers such that the
configuration reflects the hardware, and try your queries again.

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2012-07-13 17:40:57 Re: Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)
Previous Message B Sreejith 2012-07-13 16:25:20 Is there a tool available to perform Data Model review, from a performance perspective?