Re: Maximum reasonable free space map - Mailing list pgsql-general

From Grzegorz Jaśkiewicz
Subject Re: Maximum reasonable free space map
Date
Msg-id 2f4958ff0812161852r2efb3dd6n8119b5c237769e23@mail.gmail.com
Whole thread Raw
In response to Re: Maximum reasonable free space map  (Phillip Berry <pberry@stellaconcepts.com>)
List pgsql-general
On Wed, Dec 17, 2008 at 2:29 AM, Phillip Berry
<pberry@stellaconcepts.com> wrote:
>
> The data in nearly every table is constantly changing due to a high volume of new data constantly
> coming in, processing on the existing data and heavy reporting being done all at once all day and
> night.
>
> So I guess my question is, is there a point where you start to see diminishing returns or even
> negative returns by setting the fsm too high?

I personally didn't experience anything like that.
If the data is updated, but not deleted mostly - you should do
something to utilize HOT (if using 8.3).
Otherwise, you want FSM to be big enough - so that more and more
deleted records could get reused without need for so often vacuum.

--
GJ

pgsql-general by date:

Previous
From: Phillip Berry
Date:
Subject: Re: Maximum reasonable free space map
Next
From: Gregory Stark
Date:
Subject: Re: Maximum reasonable free space map