From: Scott Marlowe <scott.marlowe@gmail.com>
To: Greg Smith <gsmith@gregsmith.com>
Cc: lionel@art-informatique.com
Cc: pgsql-performance@postgresql.org
Subject: Re: Which hardware ?
Date: Tue, 17 Jun 2008 09:42:34 -0600
Message-ID: <dcc563d10806170842g6d039d7bsec81fb562ba9facd@mail.gmail.com> (raw)
In-Reply-To: <Pine.GSO.4.64.0806171108060.7150@westnet.com>
References: <g38epm$2pl4$1@news.hub.org>
<20080617142501.GG68434@commandprompt.com>
<Pine.GSO.4.64.0806171108060.7150@westnet.com>
On Tue, Jun 17, 2008 at 9:32 AM, Greg Smith <gsmith@gregsmith.com> wrote:
> Considering that Lionel's system seems pretty overpowered for what he's
> doing--runs plenty fast on a much slower system, enough RAM to hold a large
> portion of the primary tables and database, all batch updates that don't
> really need a good RAID setup--I'd say "looks good" here and recommend he
> just follow the plan he outlined. Just watch the system with top for a bit
> under load to make sure the Java processes are staying under control.
In the original post he mentioned that he had 5 years of data at about
25G / year.
With 125G of data, it's likely that if most queries are on recent data
it'll be in RAM, but anything that hits older data will NOT have that
luxury. Which is why I recommended RAID-10. It doesn't have to be on
a $1200 card with 44 disks or something, but even 4 disks in a sw
RAID-10 will be noticeably faster (about 2x) than a simple RAID-1 at
hitting that old data.
We had a reporting server with about 80G of data on a machine with 4G
ram last place I worked, and it could take it a few extra seconds to
hit the old data, but the SW RAID-10 on it made it much faster at
reporting than it would have been with a single disk.
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: scott.marlowe@gmail.com, gsmith@gregsmith.com, lionel@art-informatique.com
Subject: Re: Which hardware ?
In-Reply-To: <dcc563d10806170842g6d039d7bsec81fb562ba9facd@mail.gmail.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