From: Scott Marlowe <scott.marlowe@gmail.com>
To: Simon Riggs <simon@2ndquadrant.com>
Cc: Campbell, Lance <lance@uiuc.edu>
Cc: Heikki Linnakangas <heikki@enterprisedb.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: work_mem and shared_buffers
Date: Sun, 18 Nov 2007 08:41:59 -0600
Message-ID: <dcc563d10711180641j6dece090oebf329944f1b9599@mail.gmail.com> (raw)
In-Reply-To: <1195396179.4217.1.camel@ebony.site>
References: <B10E6810AC2A2F4EA7550D072CDE8760197DE4@SAB-FENWICK.sab.uiuc.edu>
<47349F4E.5070204@enterprisedb.com>
<B10E6810AC2A2F4EA7550D072CDE8760197DE5@SAB-FENWICK.sab.uiuc.edu>
<dcc563d10711091112r73ef016dp56d01bd6dd04a1d3@mail.gmail.com>
<1195396179.4217.1.camel@ebony.site>
On Nov 18, 2007 8:29 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
> On Fri, 2007-11-09 at 13:12 -0600, Scott Marlowe wrote:
>
> > Note that my best time was at around 16 Meg work_mem. This data set
> > is MUCH bigger than 16 Meg, it's around 300-400 Meg. But work_mem
> > optimized out at 16 Meg. Btw, I tried it going as high as 768 Meg,
> > and it was still slower than 16M.
>
> Remember that what you have shown is that for *this* dataset 16Mb is the
> optimum value. It is not a recommended value for all cases.
Actually, on this particular machine, it's held true for all the
datasets that are on it.
But I agree that it's only true for those particular datasets, and
more importantly, this machine.
But my real point was that if you haven't tested various settings on
your machine, you don't know if you're helping or hurting with various
changes to work_mem
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: scott.marlowe@gmail.com, simon@2ndquadrant.com, lance@uiuc.edu, heikki@enterprisedb.com
Subject: Re: work_mem and shared_buffers
In-Reply-To: <dcc563d10711180641j6dece090oebf329944f1b9599@mail.gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox