pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Leigh Dyer <leigh@eclinic.com.au>
To: pgsql-performance@postgresql.org
Subject: Faster drives for WAL than for data?
Date: Tue, 28 Jun 2005 11:24:48 +1000
Message-ID: <42C0A6E0.6000101@eclinic.com.au> (raw)

Hi all,

My company currently runs a number of both web-based and more 
transactional projects on a PostgreSQL 7.3 server, and we're looking to 
upgrade to a new machine running 8.0 to boost performance and handle 
data growth in to the future.

Right now I'm looking at a Sun Fire V40z server in a fairly modest 
configuration: 2 Opteron 848 (2.2ghz) CPUs, and 4GB of RAM. The V40z has 
6 drive bays, and from earlier posts and the info at 
http://www.powerpostgresql.com/PerfList/ it sounds like the best 
configuration would be:

* 2 drives in RAID 1 for OS and WAL
* 4 drives in RAID 1+0 for data

However, using 73gb 15krpm drives, I'll be limiting myself to about 
140GB of data storage, and I'm not sure if this will be enough to cover 
the life of the server. If I stick with the faster drives for the WAL, 
how significant a performance impact will there be if I use larger 
10krpm drives for the data?

Also, if anyone could recommend a SCSI RAID card for this configuration, 
or if anyone has any other suggestions, it'd be greatly appreciated.

Thanks
Leigh



Message-ID: <42C0A6E0.6000101@eclinic.com.au>
Permalink:  ../42C0A6E0.6000101@eclinic.com.au/
Also on:    postgresql.org/message-id/42C0A6E0.6000101@eclinic.com.au

 · 

reply

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: leigh@eclinic.com.au
  Subject: Re: Faster drives for WAL than for data?
  In-Reply-To: <42C0A6E0.6000101@eclinic.com.au>

* 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