Date: Mon, 12 Aug 2013 11:28:33 -0400
From: "David F. Skoll" <dfs@roaringpenguin.com>
To: pgsql-admin@postgresql.org
Subject: Opinions on SSDs
Message-ID: <20130812112833.65a97285@hydrogen.roaringpenguin.com>
Organization: Roaring Penguin Software Inc.
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk
Sender: pgsql-admin-owner@postgresql.org

Hi,

We run a fairly write-intensive workload and are looking at upgrading
our Pg servers.  (PostgreSQL 9.1; no practical way to upgrade to 9.2 for
a while because we use what's packaged with Debian.)

I'm considering the following configuration:

   Dual 4-core Intel CPU (E5620 at 2.4GHz)

   192GB of RAM

   Sixteen 240GB Intel SSD 520 series drives arranged using Linux
   RAID-10.  The RAID 10 array will use eight mirrored stripes using the
   "offset-copies" RAID10 scheme.

Questions:

1) Has anyone had experience with Intel 520 SSDs?  Are they reliable?
When they fail, do they fail nicely (ie, failure detected and bad drive
removed from RAID array) or horribly (data silently corrupted...) ?

2) Is this RAID arrangement sensible?  I've seen reports where some
people keep the pg_xlog directory on a separate RAID array, but I
don't really see the advantage with SSDs.

3) Our current workload peaks at about 5000 transactions per second;
you can assume about one-third to one-half of those are writes.  Do
you think we can get away with 16 10Krpm SATA drives instead of the
SSDs?

Regards,

David.


-- 
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin