Dean Cochrane (3/30/2012)
Berl (3/30/2012)
Although CPU time difference is not very great, the difference is much bigger for the elapsed time.
Elapsed time includes wait time and is not a good measure of performance.
Actually, it is. The wait time you speak of also includes I/O time and that can be a major part of a performance problem. I do agree, though, that you really need to take "time to display" out of the picture because it will severly mask problems. It takes a certain amount of time to display a million rows and can make differences between two methods seem much more trivial than they might be.
--Jeff Moden
Change is inevitable... Change for the better is not.