RE: optimizer time reported as?

From: John Kanagaraj <john.kanagaraj_at_hds.com>
Date: 2006-01-11 23:23:34
Message-id: BEE6A332AA61424EAE305CF89D6F75C84BADC3@USSCCEVS101.corp.hds.com

Note that in 10g, this parameter is hidden (becomes "_optimizer_max_permutations"), and the value is defaulted to 2000.

Oracle seems to have noticed that the large default value in 8i was probably incorrect. At least for Oracle's E-Biz Suite (Apps 11i), even that based on a 8i database, this value necessarily needs to be set to 2000. I had helped someone (albeit on an Apps database) debug a query that took 10 mins to parse (and less than a min to execute!)- the parameter was incorrectly set to default and the query parsed and performed under a minute.

So go ahead and satisfy our curiousity... What did sp_time (or is it sp_systime_,sql!) point to?

John Kanagaraj <
DB Soft Inc
Phone: 408-970-7002 (W)

Co-Author: Oracle Database 10g Insider Solutions http://www.amazon.com/exec/obidos/tg/detail/-/0672327910/

The opinions and facts contained in this message are entirely mine and do not reflect those of my employer or customers **

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Powell, Mark D
Sent: Wednesday, January 11, 2006 1:11 PM To: oracle-l_at_freelists.org
Subject: RE: optimizer time reported as?

I remember the default on 8.1.7 being 80,000 and Oracle support had us lower it to 2,000 while we were working on an ORA-04031 iTAR.

ora817 > @mon/parms

'For all columns: Y = Yes/True    N = No/False '
' D = Deferred,  I = Immediate,  S = System,  U = User session'
' Default indicator is unreliable if col SYS indicates chg '

Enter value for parameter: optimizer_max_permutations

                                                                    D S
S M A
                                                                    e e
y o d
NAME                            VALUE                               f s

s d j

------------------------------- ----------------------------------- - -

- - -
DESCRIPTION

optimizer_max_permutations      80000                               Y Y

N N N
optimizer maximum join permutations per query block

So how did the performance test turn out?

HTH -- Mark D Powell --

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Wolfson Larry -
lwolfs

	Sent: Wednesday, January 11, 2006 4:01 PM
	To: oracle-l_at_freelists.org
	Subject: optimizer time reported as? 
	
	
	I got asked about performance on a database that was recently

upgraded from 8.1.7.4 to 9.2.0.6.

	Looking at the initora I noticed
	optimizer_max_permutations           integer

79999

The DBA doing the upgrade said she only changed the parmaeters relating to the upgrade.

I know the default changed from 8's 80000 to 9's 2000 and there was an earlier TAR telling us to change the

80000 to 79999.

I thought this might be an issue and I ran Tim Gorman's sp_time script to see where the overhead was.

I just wanted to verify that the optimizer_max_permutations time is accounted for in the

Parsing SQL time and not somewhere else.

	    TIA
	    Larry Wolfson
	    
	
************************************************************************
*
	The information contained in this communication is confidential,
is
	intended only for the use of the recipient named above, and may
be
	legally privileged.
	
	If the reader of this message is not the intended recipient, you
are 
	hereby notified that any dissemination, distribution or copying
of this
	communication is strictly prohibited.
	
	If you have received this communication in error, please resend
this
	communication to the sender and delete the original message or
any copy
	of it from your computer system.
	
	Thank you.
	
************************************************************************

--
http://www.freelists.org/webpage/oracle-l

Received on Wed Jan 11 2006 - 23:23:34 CST