Re: distribution of the sleeps on the library cache latches

Hi Bing, I meant the case when the load (number of requests) is much higher on a particular latch comparing to over latches Regards, Ed > When you say uneven, does it mean fragmented? I am learning this too. > > > Bing > > > -Original Message- > Sent: Wednesday, October 17, 2001 4:25

Re: distribution of the sleeps on the library cache latches

Hi Steve, thanks for your reply. I'm thinking about twice increasing number of library latches ( _kgl_latch_count = 23 ) in order to mitigate loading on them. Also I would like to set _kgl_bucket_count = 8 according to output of your script. Do you think it's a good idea in my case. NAME

RE: distribution of the sleeps on the library cache latches

Hi Ed, I would agree with the _kgl_latch_count change, but the _kgl_bucket_count change seems unwarranted and extreme. Rather I suspect that the size of your library cache hash table rather reflects an oversized shared pool, probably with some use of literal SQL. @ Regards, @ Steve Adams @

Re: distribution of the sleeps on the library cache latches

ww.secularislam.org/call.htm - For Muslims > @ http://www.christianity.net.au/ - For all > > > -Original Message- > From: Edward Shevtsov [mailto:[EMAIL PROTECTED]] > Sent: Friday, 19 October 2001 18:02 > To: [EMAIL PROTECTED] > Cc: Steve Adams > Subject: Re:

RE: distribution of the sleeps on the library cache latches

For DBAs > @ http://www.secularislam.org/call.htm - For Muslims > @ http://www.christianity.net.au/ - For all > > > -Original Message- > From: Edward Shevtsov [mailto:[EMAIL PROTECTED]] > Sent: Friday, 19 October 2001 18:02 > To: [EMAIL PROTECTED] >

RE: distribution of the sleeps on the library cache latches

@ http://www.ixora.com.au/ - For DBAs > @ http://www.secularislam.org/call.htm - For Muslims > @ http://www.christianity.net.au/ - For all > > > -Original Message- > From: Edward Shevtsov [mailto:[EMAIL PROTECTED]] > Sent: Friday, 19 October 2

Re: distribution of the sleeps on the library cache latches

MAIL PROTECTED]] > Sent: Friday, 19 October 2001 19:09 > To: Steve Adams; [EMAIL PROTECTED] > Subject: Re: distribution of the sleeps on the library cache latches > > > Hi Steve, > > yes, you're absolutely right. I've inhereted that system. The > shared_pool_

Re: distribution of the sleeps on the library cache latches

Hi Ivo, we are on Linux RH, 8.1.7.0.1. I've tried cursor_sharing=force. Unfortunately, it causes ORA-600. Do 8.1.7.1(2) patches fix this problem? Regards, Ed > Hi Ed > have you tried cursor_sharing=force ? I dont know what version of oracle you > have and there are some known problems but ma

RE: distribution of the sleeps on the library cache latches

Hi Ed It seams that the big problem with force was solved 8172 [BUG:1365873] OERI:17182 / CGA corruption with CURSOR_SHARING=FORCE I havent tried it yet Ivo -Original Message- Sent: Friday, October 19, 2001 01:40 PM To: Multiple recipients of list ORACLE-L Hi Ivo, we are on Linux

Re: distribution of the sleeps on the library cache latches

@ http://www.ixora.com.au/ - For > DBAs > > @ http://www.secularislam.org/call.htm - For > Muslims > > @ http://www.christianity.net.au/ - For > all > > > > > > -----Original Message----- > > From: Edward Shevtsov [mailto:[EMAIL PROTEC

Re: distribution of the sleeps on the library cache latches

An excerpt from   PIRANHAS IN THE POOL,SQL PERFORMANCE KILLERSInvestigating the effects of literal SQL on Oracle performance   John BeresniewiczPrecise Software Solutio -- Effect of CURSOR_SHARINGA primary purpose of the 8.1.6 testing was to assess the impact the new CURSOR_S