Re: 11G Statistics Collection
From: Robert Klemme <shortcutter_at_googlemail.com>
Date: Mon, 16 Apr 2012 09:32:02 +0200
Message-ID: <9v207iFkfdU2_at_mid.individual.net>
On 14.04.2012 14:32, vsevolod afanassiev wrote:
> I think for properly written application frequent statistics
> collection isn't required. However the issue is with queries that
> depen on min/max value for a particular column. Something like
>
> SELECT * FROM TABLE WHERE CREATED_DATE> TRUNC(SYSDATE).
If it is the nature of the application that data distribution changes frequently then what other approach than frequent statistics collection on affected tables would be your solution for a "properly written application"?
Date: Mon, 16 Apr 2012 09:32:02 +0200
Message-ID: <9v207iFkfdU2_at_mid.individual.net>
On 14.04.2012 14:32, vsevolod afanassiev wrote:
> I think for properly written application frequent statistics
> collection isn't required. However the issue is with queries that
> depen on min/max value for a particular column. Something like
>
> SELECT * FROM TABLE WHERE CREATED_DATE> TRUNC(SYSDATE).
If it is the nature of the application that data distribution changes frequently then what other approach than frequent statistics collection on affected tables would be your solution for a "properly written application"?
Kind regards
robert
-- remember.guy do |as, often| as.you_can - without end http://blog.rubybestpractices.com/Received on Mon Apr 16 2012 - 02:32:02 CDT