Possible Duplicate:
How does SQL server work out the estimated number of rows?
I was just looking through SQL Server Graphical Execution plan and I encountered the following two information:
- actual number of rows-2385
- estimated number of rows-180
I was wondering why this difference was coming. Can you explain me what is the difference between them. It will be helpful if you can explain with some example.
Thanks.
Row estimates are based on statistics. Inaccurate estimates can come from a number of factors:
As a rule, don’t worry about it until you have issues.
I my experience, it’s most often based on the cardinality in your stats.
If you are selecting based on two fields, both of which have indexes, the row estimate will be based on a product of the likelihood of the values in the respective indexes, times the total number of rows.