Re: Weird performance drop after VACUUM

From: Tom Lane
Subject: Re: Weird performance drop after VACUUM
Date: ,
Msg-id: 1404.1125155101@sss.pgh.pa.us
(view: Whole thread, Raw)
In response to: Re: Weird performance drop after VACUUM  ("Steinar H. Gunderson")
List: pgsql-performance

Tree view

Weird performance drop after VACUUM  (Ümit Öztosun, )
 Re: Weird performance drop after VACUUM  (asif ali, )
  Re: Weird performance drop after VACUUM  (Philip Hallstrom, )
  Re: Weird performance drop after VACUUM  (Michael Fuhr, )
   Re: Weird performance drop after VACUUM  (asif ali, )
    Re: Weird performance drop after VACUUM  (Michael Fuhr, )
     Re: Weird performance drop after VACUUM  (asif ali, )
      Re: Weird performance drop after VACUUM  (Michael Fuhr, )
       Re: Weird performance drop after VACUUM  (asif ali, )
 Re: Weird performance drop after VACUUM  (Tom Lane, )
  Re: Weird performance drop after VACUUM  (Umit Oztosun, )
  Re: Weird performance drop after VACUUM  ("Steinar H. Gunderson", )
   Re: Weird performance drop after VACUUM  (Tom Lane, )
 Re: Weird performance drop after VACUUM  ("Steinar H. Gunderson", )
  Re: Weird performance drop after VACUUM  (Tom Lane, )

"Steinar H. Gunderson" <> writes:
> On Fri, Aug 26, 2005 at 07:31:51PM -0400, Tom Lane wrote:
>> Or you could just play with the order of the filter conditions ... for
>> example, the date condition at the end is probably far cheaper to test
>> than the text comparisons, so if that's fairly selective it'd be worth
>> putting it first.

> That's an interesting approach -- could the planner do such things itself?

It could, but it doesn't really have enough information.  We don't
currently have any model that some operators are more expensive than
others.  IIRC the only sort of reordering the current code will do
in a filter condition list is to push clauses involving sub-SELECTs
to the end.

            regards, tom lane


pgsql-performance by date:

From: Tom Lane
Date:
Subject: Re: Limit + group + join
From: Mark Kirkwood
Date:
Subject: Re: Limit + group + join