COUNT(*) or FETCH Faster?

No referential integrity constraints here. Usually, when I want to check
for records in a secondary file before I allow a delete of, say, a customer
master record, I just used SQL SELECT with COUNT(*) and return that to the
maintenance program.



Works okay, but now I need to check an invoice history table which has about
2,000,000 records, but I only want to check for the last twelve [12] months.




I was wondering if a SELECT into a cursor and then a FETCH FIRST 1 ROWS ONLY
might be faster.



I'm pretty sure the COUNT(*) option has to process all of the records for
the period I pass to it. I figured the SELECT into a cursor did, too, but
don't know for sure.



Thanks.



Jerry C. Adams

IBM i Programmer/Analyst

Try to look unimportant; the enemy may be low on ammo and not want to waste
a bullet on you.

--

A&K Wholesale

Murfreesboro, TN

615-867-5070