Re: 8.3devel slower than 8.2 under read-only load

From: Simon Riggs <simon(at)2ndquadrant(dot)com>
To: Guillaume Smet <guillaume(dot)smet(at)gmail(dot)com>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: 8.3devel slower than 8.2 under read-only load
Date: 2007-11-22 09:29:46
Message-ID: 1195723786.4246.125.camel@ebony.site
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, 2007-11-22 at 00:30 +0100, Guillaume Smet wrote:

> > Is the data identical on both systems?

Guillaume,

Sounds comprehensive, thanks for double checking.

Would it be possible to do these tests?

1. Compare SELECT 1;
This will allow us to remove planner and indexscan overheads from
results, though will still include protocol and tcop stuff.

2. Compare SELECT ... WHERE values are constants
This will cause the clients to reuse the plan already made, so should
effectively remove planner, but not indexscan overheads from the test.

3. Change the test to look at Integers columns only for the WHERE
clause, so we can remove any thought it has anything to do with text
data, collation etc..

>From those tests we should be able to narrow things down to planner,
executor or indexscan related.

Thanks,

--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Zeugswetter Andreas ADI SD 2007-11-22 09:34:52 Re: Postgres 8.3 archive_command
Previous Message Zeugswetter Andreas ADI SD 2007-11-22 09:02:03 Re: Postgres 8.3 archive_command