Re: ideal server

From: Martin Gainty <mgainty(at)hotmail(dot)com>
To: Steve Atkins <steve(at)blighty(dot)com>, pgsql-general General <pgsql-general(at)postgresql(dot)org>
Subject: Re: ideal server
Date: 2008-10-17 22:31:48
Message-ID: BLU142-W3BD22846DE6DBB2E97BF6AE320@phx.gbl
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


non-existent and/or improper DB design can impede your progress

You'll need to get stats pack configured, run explain plain on your selects to
determine which columns you will to need to index..

Martin Gainty
______________________________________________
Disclaimer and confidentiality note
Everything in this e-mail and any attachments relates to the official business of Sender. This transmission is of a confidential nature and Sender does not endorse distribution to any party other than intended recipient. Sender does not necessarily endorse content contained within this transmission.

> From: steve(at)blighty(dot)com
> To: pgsql-general(at)postgresql(dot)org
> Subject: Re: [GENERAL] ideal server
> Date: Fri, 17 Oct 2008 14:46:13 -0700
>
>
> On Oct 17, 2008, at 2:19 PM, Brian Modra wrote:
>
> > Hi,
> > I have a server in production running postgresql, receiving 110 rows
> > inserted per second, with some pretty intense queries involving long
> > plpgsql.
> > This server has 4Gigs of RAM and dual processor. Disk is Raid 5.
> > I need more power, and am wondering what is the place really I need
> > to put more emphasis? CPU, RAM, or disk?
>
> Monitoring what your performance bottlenecks are on the current system
> (after tuning postgresql reasonably) is probably the best way to get
> an answer to that for your workload.
>
> > I'm thinking of a 4xCPU and 20 Gigs and one of those large ram disks
> > which has its own battery and writes all RAM to hard disk in the
> > event of power failure.
> >
> > Obviously I need to first get a good sysadmin guy to configure
> > postgres properly, but I am sure I also need a bigger host.
> > Any advice will be appreciated.
>
> Tune first. Then monitor. Then think about what hardware changes (or
> system configuration changes) might be worth looking at.
>
> I'm betting that at that point you'll find the problem is disk I/O,
> but it may turn out not to be.
>
> RAID5 isn't considered well suited to typical database I/O, compared
> to RAID10 or similar.
>
> At the moment you have system logging, database WAL and database data
> all competing for the same I/O, meaning the drives are probably
> seeking a lot. Some ways around that are to use a decent disk
> controller with a (battery-backed) writeback cache to aggregate
> writes, or to divide your disk traffic across different disks - for
> instance, a dedicated RAID pair for the database WAL.
>
> Cheers,
> Steve
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general(at)postgresql(dot)org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general

_________________________________________________________________
You live life beyond your PC. So now Windows goes beyond your PC.
http://clk.atdmt.com/MRT/go/115298556/direct/01/

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message John Skillings 2008-10-17 22:36:01 Re: Copying Blobs between two tables using Insert stmt - [SOLUTION]
Previous Message Scott Marlowe 2008-10-17 22:16:18 Re: ideal server