Re: help tuning queries on large database

From: Tom Lane
Subject: Re: help tuning queries on large database
Date: ,
Msg-id: 23743.1136591275@sss.pgh.pa.us
(view: Whole thread, Raw)
In response to: help tuning queries on large database  (peter royal)
Responses: Re: help tuning queries on large database  (David Lang)
List: pgsql-performance

Tree view

help tuning queries on large database  (peter royal, )
 Re: help tuning queries on large database  (Tom Lane, )
  Re: help tuning queries on large database  (David Lang, )
 Re: help tuning queries on large database  (Harry Jackson, )
 Re: help tuning queries on large database  ("Luke Lonergan", )
  Re: help tuning queries on large database  (peter royal, )
   Re: help tuning queries on large database  ("Luke Lonergan", )
    Re: help tuning queries on large database  (peter royal, )
     Re: help tuning queries on large database  ("Luke Lonergan", )
   Re: help tuning queries on large database  (Ron, )
 Re: help tuning queries on large database  (Ron, )
  Re: help tuning queries on large database  (Kelly Burkhart, )
   Re: help tuning queries on large database  (Harry Jackson, )
  Re: help tuning queries on large database  (Mark Lewis, )
   Re: help tuning queries on large database  (Ron, )

peter royal <> writes:
> So, my question is, is there anything I can do to boost performance
> with what I've got, or am I in a position where the only 'fix' is
> more faster disks? I can't think of any schema/index changes that
> would help, since everything looks pretty optimal from the 'explain
> analyze' output. I'd like to get a 10x improvement when querying from
> the 'cold' state.

I don't think you have any hope of improving the "cold" state much.
The right way to think about this is not to be in the "cold" state.
Check your kernel parameters and make sure it's not set to limit
the amount of memory used for cache (I'm not actually sure if there
is such a limit on Linux, but there definitely is on some other Unixen).
Look around and see if you can reduce the memory used by processes,
or even better, offload non-database tasks to other machines.

Basically you need to get as much of the database as you can to stay
in disk cache.

            regards, tom lane


pgsql-performance by date:

From: David Lang
Date:
Subject: Re: help tuning queries on large database
From: Michael Stone
Date:
Subject: Re: improving write performance for logging