May 18, 2010 at 6:29 am
Folks, I could not find any material that answers my question.
I have a very long query in which I am using temp table. I use it because otherwise I will have to join with a very huge transaction table multiple number of times. Instead, I filter the rows into a temp table and do the join on this.
Now there is strange thing that I noticed. When I do a join with the temp table the execution plan shows the 'estimated number of rows' as 1 and the 'actual number of rows' as a bigger value.
I have read in SSC that use of UDF will cause this. But in this case I am not using a UDF. Won't the optimizer 'know' the number of rows in the temp table? The temp table is in turn derived from another temp table which is populated from the output of a UDF joined with other tables. Is this the reason?
Please see the attached execution plan. How badly could this aspect affect the performance?
I do not want to post DDL scripts as the query is quite exhaustive [but not tough :)] and not straightforward.
- confused arjun
https://sqlroadie.com/
May 18, 2010 at 7:40 am
Because there is no data in the temp table when it gets created, there are no stats. Without stats, you get an estimate of 1. You can update the stats in the temp table prior to use. That may affect the plan generated.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply