Re: Memory Tuning

From: Bruno Wolff III <bruno(at)wolff(dot)to>
To: Mitch Vincent <mitch(at)venux(dot)net>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Memory Tuning
Date: 2001-03-30 19:33:45
Message-ID: 20010330133345.A20897@wolff.to
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Fri, Mar 30, 2001 at 01:52:42PM -0500,
Mitch Vincent <mitch(at)venux(dot)net> wrote:
> If you have a specific query that you're having trouble with, post it and
> the table schema and an EXPLAIN of the query when you run it, generally
> someone will have some immediate pointers on how to speed things up.. I

I am currently interested in the general issue of how to use memory
efficiently rather than speeding up specific queries. I may revisit
that later to see if controlling join order would help.

Right now I am interested in such things such as:

Should I leave postgres tuning alone and let Linux use all of the memory
for buffer caching?

Is there any good reason to increase the number of buffers per backend over
the default of 2?

If I don't anticipate a lot of simultaneous queries, what fraction of memory
should I let in core sorts take up?

To set the estimate for bucher caching, should I just look at /proc/meminfo
to see how much memory is being used for caching or should I get this
estimate some other way?

Why I am not seeing consitant wall clock times for queries? Presumably
there is some caching going on, but I am not sure if it is in postgres
or in the OS.

> don't have time to go through your site looking for the database schema and
> such but if you include some specific information in an email to the list

Including the pointers to the site was a semi joke. I don't expect people
to go and give me specific answers. And I am not as interested in that
as I am in general ideas in how to figure things out.

This might include useful rules of thumb for (for my current issue) memory
tuning, where to read about performance measuring tools.

> I'd be happy to take a quick look (and I'm sure others would too)..

I appreciate the help, but I think so far you are offering to answer a
different question than I have at this point.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Doug McNaught 2001-03-30 19:39:50 Re: Globally Unique IDs?
Previous Message Mike Mascari 2001-03-30 19:20:22 RE: Globally Unique IDs?