We had another deadlock on our sql server. I found that two users, independently, were UPDATing the same table and one was chosen as the deadlock victim (One had a X lock on the table, the other had a U lock but wanted an X). .Each UPDATE statement's WHERE clause share some table columns in common and each also uses unique columns. How do I determine which columns to use for a new non-clustered index?
TIA,
edm2