From: Justin Clift <justin@postgresql.org>
To: Don Baccus <dhogaza@pacifier.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@postgresql.org
Subject: Re: Increasing checkpoint distance helps 7.2 noticeably
Date: Thu, 10 Jan 2002 20:16:25 +1100
Message-ID: <3C3D5BE9.EA3504A5@postgresql.org> (raw)
References: <17608.1010631664@sss.pgh.pa.us>
<3C3D0A2E.5020501@pacifier.com>
Don Baccus wrote:
>
> Tom Lane wrote:
>
> > BTW, has anyone experimented with the OSDB benchmark at
> > http://osdb.sourceforge.net/ ? I'm wondering if it might give
> > more useful numbers than pgbench does.
I heartily recommend OSDB. It seems to give reliable numbers, and
simulates multiple clients quite well.
It's the continuation of the code which Compaq sponsored for internal
testing of their own servers.
It's based on the AS3AP (ANSI SQL, Standard Scalable and Portable)
benchmark, with the exception it generates a lot more information about
how the database is doing, unlike the official benchmark which pretty
much just generates one number at the end.
The AS3AP benchmark is described and documented at :
http://www.benchmarkresources.com/handbook/5.html
Version 0.11 of the OSDB software
(http://sourceforge.net/project/showfiles.php?group_id=18681&release_id=55146)
supports PostgreSQL 7.1+.
I remember having trouble with hash indices, and having to modify the
code to use btree indices instead, as PostgreSQL seems to have
deadlocking problems with hash indices when multiple people access them.
Also, you have to generate your own dataset using a specific program (I
have it if anyone needs it), or download the test datasets. The test
datasets have a maximum of about 40MB, but if you use the program to
generate your own data, you can generate test data up to about 44GB or
so (or maybe more, I don't remember).
Hope that info is useful for someone.
:-)
Regards and best wishes,
Justin Clift
>
> Graphs like this are worth saving and adding to the documentation as an
> aid to understanding various approaches to tuning.
>
> Along with the scripts that generate the graphs as an example that folks
> can use if they have to tune a particular set of queries on a particular
> set of data.
>
> --
> Don Baccus
> Portland, OR
> http://donb.photo.net, http://birdnotes.net, http://openacs.org
>
> ---------------------------(end of broadcast)---------------------------
> TIP 3: if posting/reading through Usenet, please send an appropriate
> subscribe-nomail command to majordomo@postgresql.org so that your
> message can get through to the mailing list cleanly
--
"My grandfather once told me that there are two kinds of people: those
who work and those who take the credit. He told me to try to be in the
first group; there was less competition there."
- Indira Gandhi
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-hackers@postgresql.org
Cc: justin@postgresql.org, dhogaza@pacifier.com, tgl@sss.pgh.pa.us
Subject: Re: Increasing checkpoint distance helps 7.2 noticeably
In-Reply-To: <3C3D5BE9.EA3504A5@postgresql.org>
* 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