Re: Scaling further up

From: "Magnus Hagander" <mha(at)sollentuna(dot)net>
To: "Anjan Dave" <adave(at)vantage(dot)com>, <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Scaling further up
Date: 2004-03-01 20:54:20
Message-ID: 6BCB9D8A16AC4241919521715F4D8BCE34B351@algol.sollentuna.se
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

> All:
>
> We have a Quad-Intel XEON 2.0GHz (1MB cache), 12GB memory, running
RH9, PG 7.4.0. There's
> an internal U320, 10K RPM RAID-10 setup on 4 drives.
>
> We are expecting a pretty high load, a few thousands of 'concurrent'
users executing either
> select, insert, update, statments.

> What is the next step up in terms of handling very heavy loads?
Clustering?

I'd look at adding more disks first. Depending on what type of query
load you get, that box sounds like it will be very much I/O bound. More
spindles = more parallell operations = faster under load. Consider
adding 15KRPM disks as well, they're not all that much more expensive,
and should give you better performance than 10KRPM.

Also, make sure you put your WAL disks on a separate RAIDset if possible
(not just a separate partition on existing RAIDset).

Finally, if you don't already have it, look for a battery-backed RAID
controller that can do writeback-cacheing, and enable that. (Don't even
think about enabling it unless it's battery backed!) And add as much RAM
as you can to that controller.

//Magnus

Browse pgsql-performance by date

  From Date Subject
Next Message Anjan Dave 2004-03-01 21:02:27 Re: Scaling further up
Previous Message Rob Fielding 2004-03-01 17:30:41 Re: WAL Optimisation - configuration and usage