User-Agent: Microsoft-Entourage/11.2.1.051004
Date: Wed, 01 Feb 2006 20:47:09 -0800
Subject: Re: Huge Data sets, simple queries
From: "Luke Lonergan" <llonergan@greenplum.com>
To: "Jeffrey W. Baker" <jwbaker@acm.org>,
	"Lance Lierheimer" <llierheimer@greenplum.com>
cc: pgsql-performance@postgresql.org
Message-ID: <C006CECD.1B7FF%llonergan@greenplum.com>
Thread-Topic: [PERFORM] Huge Data sets, simple queries
Thread-Index: AcYnCQ7Mv1XVKnpYSsuSoYyFtgu+SwAqqqcy
In-Reply-To: <1138782313.14732.1.camel@noodles>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=us-ascii
Content-Transfer-Encoding: 7bit

Jeffrey,

On 2/1/06 12:25 AM, "Jeffrey W. Baker" <jwbaker@acm.org> wrote:

> Ah, but someday Pg will be able to concurrently read from two
> datastreams to complete a single query.  And that day will be glorious
> and fine, and you'll want as much disk concurrency as you can get your
> hands on.

Well - so happens that we have one of those multi-processing postgres'
handy, so we'll test this theory out in the next couple of days.  We've a
customer who ordered 3 machines with 6 drives each (Dell 2850s) on two U320
SCSI busses, and we're going to try configuring them all in a single RAID10
and run two Bizgres MPP segments on that (along with two mirrors).

We'll try the RAID10 config and if we get full parallelism, we'll use it (if
the customer like it).  Otherwise, we'll use two 3 disk RAID5 sets.

I'll post the results here.

Thanks Jeffrey,

- Luke