oh Cary/Dan/Jared thank you.... it's been a rough day and this made me laugh out loud
--- Cary Millsap <[EMAIL PROTECTED]> wrote: > From another session, it's: > > exec dbms_system.set_imaginary_param_in_session( > $SID, > $SERIAL, > 'sql_execution_speed', > 'warp_10' > ) > > > Cary Millsap > Hotsos Enterprises, Ltd. > http://www.hotsos.com > > Upcoming events: > - Hotsos Clinic, Oct 1-3 San Francisco, Oct 15-17 Dallas, Dec 9-11 > Honolulu > - 2003 Hotsos Symposium on OracleR System Performance, Feb 9-12 > Dallas > - Next event: NCOAUG Training Day, Aug 16 Chicago > > > > -----Original Message----- > [EMAIL PROTECTED] > Sent: Wednesday, August 14, 2002 3:43 PM > To: Multiple recipients of list ORACLE-L > > Oh, in that case: > > ALTER SYSTEM SET SQL_EXECUTION_SPEED=WARP_10; > > ;) > > Jared > > > > > > "Khedr, Waleed" <[EMAIL PROTECTED]> > Sent by: [EMAIL PROTECTED] > 08/14/2002 08:58 AM > Please respond to ORACLE-L > > > To: Multiple recipients of list ORACLE-L > <[EMAIL PROTECTED]> > cc: > Subject: RE: It took more than an hour to update > 10,000 > records > > > I think the original poster was testing our skills since he posted > his > problem and disappeared. > Also he asked for some init.ora parameter to make it faster:) > > Regards, > > Waleed > > > > -----Original Message----- > To: [EMAIL PROTECTED]; Khedr, Waleed > Sent: 8/14/02 10:42 AM > > > It wasn't 20 seconds it was 15 seconds. > > The same lessons apply. Without knowing what else > is going on in the database, it's not possible to say why > it's taking an hour to insert 10k rows. > > Using bind variables will drastically reduce the resources > consumed by the insert job. > > Using SQL*loader with a generous bindsize argument would > be even better. > > Running the queries on the wait interface is a start to learning > what the problem is. > > But then, I'm sure you're already aware of all this, and I'm > no doubt preaching to the choir. > > Jared > > On Tuesday 13 August 2002 18:43, Khedr, Waleed wrote: > > His problem is not reducing 60 sec to 20 seconds but why an hour or > more > > was taken to update 10,000 rows. > > > > It's obvious there are other things (which I indicated in my email) > that > > need to be sorted out to reduce the hour to few minutes. > > > > Regards, > > > > Waleed > > > > -----Original Message----- > > Sent: Tuesday, August 13, 2002 9:08 PM > > To: Multiple recipients of list ORACLE-L > > > > > > David, > > > > You don't yet know why it's slow. > > > > Why not turn tracing on for the session doing the inserts and then > > run tkprof on the trace file? > > > > Since the inserts seem to be taking so long, you should also > > be able to join v$session_wait with v$session to catch the > > waits in action and determine exactly what they are. > > > > You should also join v$session and v$session_event while > > the insert job is running. > > > > Actually, using bind variables will probably help. The insert > > statements will be less resource intensive and run much faster, > > and this will result in less contention for other sessions. > > > > My tests of insert statements with and without bind variables > > show that using bind variables resulted in an insert job > > of 27k rows running in 15 seconds and < 1/100 sec of parse time. > > > > Not using bind variables required 60 seconds, 20 of which were > > CPU parse time. > > > > The increased run time also resulted in 3x as many buffer busy > > waits for other sessions trying to hit the same table. > > > > HTH > > > > Jared > > > > > > > > > > > > > > > > > > "Nguyen, David M" <[EMAIL PROTECTED]> > > Sent by: [EMAIL PROTECTED] > > 08/13/2002 01:34 PM > > Please respond to ORACLE-L > > > > > > To: Multiple recipients of list ORACLE-L > <[EMAIL PROTECTED]> > > cc: > > Subject: RE: It took more than an hour to update > 10,000 > > records > > > > > > Is there a parameter that I can tune in init.ora file to improve > this > > process? > > > > Thanks, > > David > > > > -----Original Message----- > > Sent: Tuesday, August 13, 2002 2:00 PM > > To: Multiple recipients of list ORACLE-L > > > > > > Bind variables will not buy you anything. The time taken is mainly > > dependent > > on the execution plan and the cardinality of the indexes being > used. > Also > > any overhead like updating a column that's part of an index, > triggers, > > referential constraints, etc. > > > > Waleed > > > > -----Original Message----- > > Sent: Tuesday, August 13, 2002 2:35 PM > > To: Multiple recipients of list ORACLE-L > > > > > > I run it on the server. How do I use bind variables? > > > > Thanks, > > David > > > > -----Original Message----- > > Sent: Tuesday, August 13, 2002 1:13 PM > > To: Multiple recipients of list ORACLE-L > > > > > > From the look of your sql statements it seems that each > > statement will be hard parsed because the values are different. > === message truncated === __________________________________________________ Do You Yahoo!? HotJobs - Search Thousands of New Jobs http://www.hotjobs.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Rachel Carmichael INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
