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

From Claudio Freire
Subject Re: Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)
Date
Msg-id CAGTBQpaQ066-vUZ_eN3CoND2W4QfL_tDoxYtGv87Lu27DdjobA@mail.gmail.com
Whole thread Raw
In response to Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)  (Eoghan Murray <eoghan@qatano.com>)
List pgsql-performance
On Fri, Jul 13, 2012 at 11:11 AM, Eoghan Murray <eoghan@qatano.com> wrote:
> 8.4: 314ms:  http://explain.depesz.com/s/GkX
> 9.1: 10,059ms :http://explain.depesz.com/s/txn
> 9.1 with setting `enable_material = off`: 1,635ms
> http://explain.depesz.com/s/gIu

I think the problem is it's using a merge join, with a sort inside
that's producing 600x more rows than expected, while 8.4 does a hash
join with no intermediate big tables instead.

What's your configuration like in both servers? (that could explain
planning differences)

pgsql-performance by date:

Previous
From: Eoghan Murray
Date:
Subject: Poor performance problem with Materialize, 8.4 -> 9.1 (enable_material)
Next
From: Tom Lane
Date:
Subject: Re: query overhead