SQL Distinct keyword bogs down performance?

Question

I have received a SQL query that makes use of the distinct keyword. When I tried running the query it took at least a minute to join two tables with hundreds of thousands of

Answer 1

Distinct always sets off alarm bells to me - it usually signifies a bad table design or a developer who's unsure of themselves. It is used to remove duplicate rows, but if the joins are correct, it should rarely be needed. And yes there is a large cost to using it.

What's the primary key of the orders table? Assuming it's orderno then that should be sufficient to guarantee no duplicates. If it's something else, then you may need to do a bit more with the query, but you should make it a goal to remove those distincts! ;-)

Also you mentioned the query was taking a while to run when you were checking the number of rows - it can often be quicker to wrap the entire query in "select count(*) from ( )" especially if you're getting large quantities of rows returned. Just while you're testing obviously. ;-)

Finally, make sure you have indexed the custID on the orders table (and maybe recDate too).