Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: time attributable to FGAC rewrite

Re: time attributable to FGAC rewrite

From: robert <gnuoytr_at_rcn.com>
Date: 14 Jul 2004 18:00:15 -0700
Message-ID: <da3c2186.0407141700.5f06771d@posting.google.com>


"Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in message news:<cd3usm$nq5$1_at_sparta.btinternet.com>...
> Note in-line.

ditto

>
> --
> Regards
>
> Jonathan Lewis
>
> http://www.jlcomp.demon.co.uk
>
> http://www.jlcomp.demon.co.uk/faq/ind_faq.html
> The Co-operative Oracle Users' FAQ
>
> http://www.jlcomp.demon.co.uk/seminar.html
> Optimising Oracle Seminar - schedule updated May 1st
>
>
> "Daniel Morgan" <damorgan_at_x.washington.edu> wrote in message
> news:1089765107.12149_at_yasure...
> > robert wrote:
> >
> > > been looking through this group, and what docs i have, but can't
> > > find an estimate/definitive measure of how much (additional) time
> > > to execute an arbitrary query might be attributable to FGAC.
> > >
> >
> > No blanket percent increase is possible as what happens depends on
> > what you've written. But from my experience the difference is small
> > enough to be insignificant in most situations.
> >
>
> I'm interested in the 'insignificant'.

i guess i am, too.

 The version 9 implementation
> of FGAC results in an anonymous block being executed to call
> the predicate function

on every execution of a query that contains
> a table that is subject to a policy.

my reading of Mr. Kyte's article and Mr. Feuerstein's book is that the policy function is inserted as part of the "regular" parsing exercise, not as an additional step or additionally executed. any query caching (not familiar enough with oracle jargon) still happens. at least 8i. that is, the WHERE is clause is modified only once. i admit, i'm not sure at all how often 'once' happens.

 I would have expected this to
> be quite expensive on a busy OLTP system - though probably
> ignorable on a DW/DSS system.
>
> Can you give us some idea of the scale and type of activity of
> the systems where you've tested it.
>
> >
> > Daniel Morgan
> >
Received on Wed Jul 14 2004 - 20:00:15 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US