From: John A Meinel <john@arbash-meinel.com>
To: David Parker <dparker@tazznetworks.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: seqential vs random io
Date: Mon, 23 May 2005 17:23:41 -0500
Message-ID: <429257ED.6020004@arbash-meinel.com> (raw)
In-Reply-To: <07FDEE0ED7455A48AC42AC2070EDFF7C746598@corpsrv2.tazznetworks.com>
References: <07FDEE0ED7455A48AC42AC2070EDFF7C746598@corpsrv2.tazznetworks.com>
David Parker wrote:
> I just got a question from one our QA guys who is configuring a RAID 10
> disk that is destined to hold a postgresql database. The disk
> configuration procedure is asking him if he wants to optimize for
> sequential or random access. My first thought is that random is what we
> would want, but then I started wondering if it's not that simple, and my
> knowledge of stuff at the hardware level is, well, limited.....
>
> If it were your QA guy, what would you tell him?
>
> - DAP
Random. Sequential is always pretty fast, it is random that hurts.
The only time I would say sequential is if you were planning on
streaming large files (like iso images) with low load.
But for a DB, even a sequential scan will probably not be that much data.
At least, that's my 2c.
John
=:->
Attachments:
signature.asc (application/pgp-signature, 250B) download
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: john@arbash-meinel.com, dparker@tazznetworks.com
Subject: Re: seqential vs random io
In-Reply-To: <429257ED.6020004@arbash-meinel.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