RE: Oracle 911 Article

I don't mind sharing.

session_cursorsor_cache.sql is available at ixora.com.au

The number of cacheable cursors is greater today than
when I ran tests to get an approximate value.

Sorry, no graphs and I don't seem to have save the output
of previous tests.  S_C_C was at the default of 0, so no cursor
sharing was being done.

Hard parses have decreased quite a bit.

I can see how parse time has changed after it has run 
a few days.  ( statspack every 15 minutes )


12:53:43 CIMORAPROD - jkstill@dal3 SQL> @session_cursor_cache

PARAMETER                     VALUE USAGE
----------------------------- ----- -----
session_cached_cursors         1500  100%
open_cursors                    100 ####%

CURSOR_CACHE_HITS SOFT_PARSES HARD_PARSES
----------------- ----------- -----------
           85.15%       9.82%       5.03%

MAX_CACHEABLE_CURSORS
---------------------
                 2160
12:53:49 CIMORAPROD - jkstill@dal3 SQL>






Paul Drake <discgolfdba@xxxxxxxxx>
Sent by: oracle-l-bounce@xxxxxxxxxxxxx
 03/05/2004 12:24 PM
 Please respond to oracle-l

 
        To:     oracle-l@xxxxxxxxxxxxx
        cc: 
        Subject:        RE: Oracle 911 Article


--- Jared Still <jkstill@xxxxxxxxxx> wrote:
> We have a similar change management process.
> 
> I am 'privileged' to be the owner of that process. 
> :)
> 
> Finally got to bounce the box yesterday where I had
> set session_cached_cursor=1500 over a month ago.
> 
> And yes, it was tested first, and the results
> shared.
> 
> Unless there is an absolute emergency due to a
> downed
> system or severely impacted performance, changes go
> through a review meeting first.
> 
> Emergency changes are still discussed and recorded.
> 
> For many of you, wait will Sarbanes Oxley is gets
> hold
> of you.  You will either be implementing change
> control,
> or vastly 'improving' your current process.
> 
> Jared

session_cached_cursor=1500

Really?
I'd love to see the data (or graph) of parse reduction
(and associated resource reduction) vs.
session_cached_cursors.

I just hiked it to 100 and left it there.

Hmmm. Guess I'm interested in "The Burden of Proof" if
you care to share.

thanks.

Pd

> 
> 
> On Fri, 2004-03-05 at 01:51, Niall Litchfield wrote:
> > > You guys are going to enjoy this :-).
> > > 
> > > I just read a very interesting article that
> managed to insult 
> > > some of my
> > > favorite Oracle authors and the entire state of
> Alabama!
> > 
> > Hmm we have a script with a number of ctas
> statements in it
> > 
> > Where 'pope'='catholic'
> > Where 'manager name'='in'
> > Where 'niall'='big and clever' etc
> > 
> > I read the alabama comment in the same light. 
> > 
> > >  Now 
> > > I've heard Mr.
> > > Burleson speak before, and I must admit that I
> thought he was 
> > > a dynamic and
> > > entertaining speaker.  But do you really think
> that Oracle 
> > > Support has ever
> > > told anyone to call HIM to solve their
> performance problems 
> > > (his article
> > > makes this claim)?
> > 
> > I doubt that oracle support make specific partner
> recommendation in an iTAR though they might. On the
> other hand if you have purchased support from oracle
> because you have no-one experienced in house (though
> why you would then be running oracle is beyond me) I
> can imagine support saying something along the lines
> of , "well if you are uncomfortable with doing this
> you might hire either ourselves or a third party
> oracle specialist to do this work for you". The
> first consultancy site that is returned in a google
> search for "oracle expert" is www.dba-oracle.com
> > 
> > In general I didn't like the tone of the article,
> but I do have to agree with jonathan that the
> article focusses on business problems and
> implications for clients, I like that. That said I
> find this approach hard to square with "Unlike
> academic and scientific Oracle DBAs who demand
> proofs and reproducible results before making an
> Oracle change". We can't be the only Oracle house
> that has a change management process that involves
> reproducing problems  on test and applying a fix to
> show that it works first can we? This is a business
> process and definitely not a result of the DBA
> insisting on proof before changes are applied. All
> good change management requires some evidence that
> the change will have the desired effect - doesn't
> it? 
> > 
> > Niall Litchfield
> > Oracle DBA
> > Audit Commission
> > +44 117 975 7805 
> > 
> > 
> > 
> > 
> > 
> > 
> >
>
**********************************************************************
> > This email contains information intended for
> > the addressee only.  It may be confidential
> > and may be the subject of legal and/or
> > professional privilege.  Any dissemination,
> > distribution, copyright or use of this
> > communication without prior permission of
> > the sender is strictly prohibited.
> >
>
**********************************************************************
> > 
> >
>
----------------------------------------------------------------
> > Please see the official ORACLE-L FAQ:
> http://www.orafaq.com
> >
>
----------------------------------------------------------------
> > To unsubscribe send email to: 
> oracle-l-request@xxxxxxxxxxxxx
> > put 'unsubscribe' in the subject line.
> > --
> > Archives are at
> http://www.freelists.org/archives/oracle-l/
> > FAQ is at
> http://www.freelists.org/help/fom-serve/cache/1.html
> >
>
-----------------------------------------------------------------
> 
>
----------------------------------------------------------------
> Please see the official ORACLE-L FAQ:
> http://www.orafaq.com
>
----------------------------------------------------------------
> To unsubscribe send email to: 
> oracle-l-request@xxxxxxxxxxxxx
> put 'unsubscribe' in the subject line.
> --
> Archives are at
> http://www.freelists.org/archives/oracle-l/
> FAQ is at
> http://www.freelists.org/help/fom-serve/cache/1.html
>
-----------------------------------------------------------------


__________________________________
Do you Yahoo!?
Yahoo! Search - Find what you're looking for faster
http://search.yahoo.com
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

Other related posts: