tag:blogger.com,1999:blog-518481768015386858.post607341131592409960..comments2023-08-06T04:28:52.092-04:00Comments on Karen Morton: Plan execution statisticsKarenhttp://www.blogger.com/profile/03309823327597536648noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-518481768015386858.post-54002388549290057002008-07-11T18:02:00.000-04:002008-07-11T18:02:00.000-04:00Thanks for the info Kurt. I'll certainly look int...Thanks for the info Kurt. I'll certainly look into it in more depth.<BR/><BR/>If it's this parameter that is making the rowsource stats to not add up properly to the top parent, I guess I still don't understand why they sample at all since it adds no value (as far as I can tell). It seems more confusing than anything.<BR/><BR/>Again, thanks!Karenhttps://www.blogger.com/profile/03309823327597536648noreply@blogger.comtag:blogger.com,1999:blog-518481768015386858.post-62382849906531042992008-07-09T21:21:00.000-04:002008-07-09T21:21:00.000-04:00Hi Karen. Sorry I didn't make it to the Symposium...Hi Karen. Sorry I didn't make it to the Symposium this past Spring for our annual visit.<BR/><BR/>I believe you will find a lot of answers on the statistics inconsistency between parent and child row source stats if you do some research on the hidden parameter _rowsource_statistics_sampfreq. While you might not find to much official documentation on it, both Jonathan and Julian have some content on the topic. In short, the statistics are gathered based on sampling the row source operations. This parameter controls the granularity of the sampling. Lowering the granularity can have an adverse effect on the performance of statements, so handle with care.<BR/>KurtKurt Grausteinhttps://www.blogger.com/profile/01580896621506804233noreply@blogger.com