Skip site navigation (1) Skip section navigation (2)

Re: Increasing statistics results in worse estimates

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Shelby Cain <alyandon(at)yahoo(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Increasing statistics results in worse estimates
Date: 2005-04-29 17:26:04
Message-ID: 7127.1114795564@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-generalpgsql-hackers-win32
Shelby Cain <alyandon(at)yahoo(dot)com> writes:
> I had already removed proprietary data to try and
> whittle down the number of columns I needed to
> demonstrate the weirdness so I can host a dump of the
> table.  However, before I take that step I should
> mention that this is the native Windows port so if
> that changes anything let me know.

Well, if I can't reproduce the misbehavior on a Unix machine then
there'll be reason to suspect a platform-specific bug ... but it's
still a bug.

It'll be important to know the exact datatypes of the columns,
as well as the database locale and encoding you are using.

			regards, tom lane

In response to

pgsql-hackers-win32 by date

Next:From: Tom LaneDate: 2005-05-01 17:03:55
Subject: Re: [GENERAL] Increasing statistics results in worse estimates
Previous:From: Shelby CainDate: 2005-04-29 16:54:11
Subject: Re: Increasing statistics results in worse estimates

pgsql-general by date

Next:From: Mike NolanDate: 2005-04-29 17:32:26
Subject: Tuning queries inside a function
Previous:From: Tom LaneDate: 2005-04-29 17:23:53
Subject: Re: Composite types as columns used in production?

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group