Thanks for the pointer Carol. I was reading up on the ML note and I saw this: "This parameter is generally overly restrictive in what it actually allows to be shared. SIMILAR tells oracle to try and share cursors by replacing all literals with binds for legacy applications, but directs that sharing only be performed when all the replaced literal values were exactly the same (in the case of predicates referencing columns with histograms or using inequality operators such as BETWEEN, <, and !=) " I am confused by what "but directs that sharing only be performed when all the replaced literal values were exactly the same" means. I thought SIMILAR would make Oracle use the already existing plan without having to reparse in case of **different** literal values. Is that the right understanding of 'sharing of the cursor' ? If sharing is performed only when replaced literal values are the exactly same, wouldnt that make it EXACT and not SIMILAR? Thanks. 2010/11/25 <Laimutis.Nedzinskas@xxxxxx> > nice, one more AI decoration gone. > > > > --------------------------------------------------------------------------------- > > Please consider the environment before printing this e-mail > > > > Carol Dacko > <dackoc@xxxxxxxxx > > To > Sent by: Oracle-L Freelists > oracle-l-bounce@f <oracle-l@xxxxxxxxxxxxx> > reelists.org cc > > Subject > 2010.11.23 18:49 Thank you Oracle Support - > cursor_sharing = similiar is > deprecated! > Please respond to > dackoc@xxxxxxxxx > > > > > > > > > Maybe I am the last person to the know but Doc ID: 1169017.1 is the > announcement that cursor_sharing = similar is now deprecated in 11g! And > the write up is very good. And it is great that they also included in the > references the Optimizer Team's blog about it too. > > So yes, thank you Oracle Support! I really do appreciate this! This will > now be one more battle I don't need to fight! > > Warm regards, > Carol Dacko > University of Michigan > > > > -- > //www.freelists.org/webpage/oracle-l > > >