Re: PostgreSQL win32 fragmentation issue

From: Shane Ambler <pgsql(at)007Marketing(dot)com>
To: "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com>
Cc: Andrew Dunstan <andrew(at)dunslane(dot)net>, Martijn van Oosterhout <kleptog(at)svana(dot)org>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: PostgreSQL win32 fragmentation issue
Date: 2006-12-03 03:03:59
Message-ID: 45723E9F.2080307@007Marketing.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Joshua D. Drake wrote:

> I am trying to get as much information as possible so that:
>

Well I have a small postgresql setup on my girlfriends Win XP Pro
machine that only had a little testing use a few months ago.

The data folder is 84MB and even though there is 25% total fragmentation
and 45% file fragmentation on the drive - the largest being the
pc-cillin log with 75,104 fragments - none of the fragmented files are
postgresql data files. (Been a while since I defraged this)

If we stop and think about how windows (and mac os x) allocate file
space as compared to *nix file systems then we can assume a few things
that could be tested to verify.

The first test you mentioned you defraged with an existing db then ran
the tests and then checked fragmentation again. Try this instead -
remove the contents of the data folder - defrag - initdb - run test and
see what fragmentation you get.

Try installing the data folder on a separate drive with nothing else on
it and run the test and then look at fragmentation.

How about some performance tests to measure the difference between badly
fragmented data files and un-fragmented data files. It may not make a
significant difference.

--

Shane Ambler
pgSQL(at)007Marketing(dot)com

Get Sheeky @ http://Sheeky.Biz

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2006-12-03 03:51:57 Re: GUC description cleanup
Previous Message Thomas H. 2006-12-02 23:58:46 Re: PostgreSQL win32 fragmentation issue