Dave Ballantyne (7/2/2009)
Scalar udfs are generally quite slow and should be avoided.Try to isolated the issue , what is performance like if you make it a sequential count ?
Also i notice that code is used is the clustered primary key , it could be page splitting on the inserts ?
Ouch, you can pretty much be sure there are page splits happening in a million row insert.
Like Gail has said, without knowing what the code is in the function there really isn't a way to make it faster because the UDF will still run a million times.
Can you at least give us some pseudocode for the UDF like:
Take the second parameter passed in
Loop through it by character, multiplying the the ASCII Code by the first parameter and if > 255 substract 255 and place that character in the 10 character string.
Jack Corbett
Consultant - Straight Path Solutions
Check out these links on how to get faster and more accurate answers:
Forum Etiquette: How to post data/code on a forum to get the best help
Need an Answer? Actually, No ... You Need a Question