Instead of "SIZE SKEWONLY", I would use "SIZE AUTO". It collects the same histograms as skewonly, but while skewonly will create a histogram for any column which is skewed enough, "auto" will do it only for columns which may benefit from a histogram, i.e. columns used in predicates. Of course, I prefer not to use either, but collect histograms for selected columns with individualized numbers of buckets. At 09:42 AM 8/13/2004, you wrote: >Then all the useres started to complaint somep rocess were slower, some of >them were 100% slower, then I gather againt histogram statistics and all >return to normality. >EXEC DBMS_STATS.GATHER_database_STATS( ESTIMATE_PERCENT=>100,METHOD_OPT=> >FOR ALL COLUMNS SIZE SKEWONLY'); > >So based on my experience in 9.2, if you don't use histograms, try to run >this command and see if you get some performance (test it ) > >Juan Carlos Reyes Pacheco >OCP Regards Wolfgang Breitling Centrex Consulting Corporation www.centrexcc.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 //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------