From: Carlos Moreno <moreno_pg@mochima.com>
To: Daniel Griscom <griscom@suitable.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Throttling PostgreSQL's CPU usage
Date: Wed, 09 May 2007 10:29:56 -0400
Message-ID: <4641DAE4.4020805@mochima.com> (raw)
In-Reply-To: <p06240802c2677e4804b2@[192.168.1.8]>
References: <20070509055655.D4C24DCC04A@svr2.hagander.net>
<p06240802c2677e4804b2@[192.168.1.8]>
Daniel Griscom wrote:
> Thanks again for all the feedback. Running on a dual processor/core
> machine is clearly a first step, and I'll look into the other
> suggestions as well.
As per one of the last suggestions, do consider as well putting a dual
hard disk
(as in, independent hard disks, to allow for simultaneous access to
both). That
way, you can send the WAL (px_log directory) to a separate physical drive
and improve performance (reduce the potential for bottlenecks) if there
is a
considerable amount of writes. With Windows, you can mount specific
directories to given HD partitions --- that would do the trick.
Also, of course, do make sure that you give it a generous amount of RAM, so
that an as-large-as-possible fraction of the read operations are done
directly
off the machine's memory.
BTW, have you considered using a *separate* machine for PostgreSQL?
(that way this machine could be running on Linux or some Unix flavor,
and the Windows machine is dedicated to the ActiveX rendering stuff).
I mean, if you are going to get a new machine because you need to replace
it, you might as well get a new machine not as powerful, since now you
will have the dual-CPU given by the fact that there are two machines.
Good luck!
Carlos
--
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: moreno_pg@mochima.com, griscom@suitable.com
Subject: Re: Throttling PostgreSQL's CPU usage
In-Reply-To: <4641DAE4.4020805@mochima.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