Re: RAC partitioning question
From: Marcin Przepiorowski <pioro1_at_gmail.com>
Date: Wed, 1 Feb 2012 13:24:10 +0000
Message-ID: <CAGdek=yTstqCAK5TwnmevQcQaasikca+UrYp26rWwK9=0rsm5Q_at_mail.gmail.com>
On Wed, Feb 1, 2012 at 11:40 AM, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk> wrote:
>
> Marcin,
>
> I've had a look at the blog item, and don't think it's really making a
> point that's restricted to partitioning.
>
> Any time you have mechanisms that only ever add data at the right hand side
> of an index, and then have bulk deletes scattered randomly through the
> entire length of the index, the utilisation of the average leaf block is
> bound to drop slowly over a number of delete cycles. This will happen
> whether or not you're talking about partitioning.
Date: Wed, 1 Feb 2012 13:24:10 +0000
Message-ID: <CAGdek=yTstqCAK5TwnmevQcQaasikca+UrYp26rWwK9=0rsm5Q_at_mail.gmail.com>
On Wed, Feb 1, 2012 at 11:40 AM, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk> wrote:
>
> Marcin,
>
> I've had a look at the blog item, and don't think it's really making a
> point that's restricted to partitioning.
>
> Any time you have mechanisms that only ever add data at the right hand side
> of an index, and then have bulk deletes scattered randomly through the
> entire length of the index, the utilisation of the average leaf block is
> bound to drop slowly over a number of delete cycles. This will happen
> whether or not you're talking about partitioning.
Hi Jonathan,
Thanks for you comment.
I realize that this is not restricted to partitioning only but I found
out that during partition drop index is growing faster - even with
same insert rate.
I'm going to check it again and use you script as well.
-- Marcin Przepiorowski http://oracleprof.blogspot.com -- http://www.freelists.org/webpage/oracle-lReceived on Wed Feb 01 2012 - 07:24:10 CST