Re: A kind of... benchmark?

I agree with your points on benchmarks, but I do have an exception. It's true that benchmarks like TPC, et. al. are more for marketing that anything else. After all, who would spend that kind of money on that kind of configuration to get those performance numbers? The companies that set those benchmarks surely didn't spend that kind of money. If Oracle and HP team up to provide a system that sets the latest TPC record, do you think that Oracle and HP spent all that money on the system? Not at all. It is you and I that paid for it by purchasing Oracle and HP products...but I digress.

I do use benchmarks but only for similar systems and to get a *rough* idea of how things are different. For instance, we had a db server that had one vendors disks attached to it. We purchased another vendors disks and wanted to see how they compared. So we ran some benchmark tests and we calculated different metrics to compare performance. This just gave us a rough idea. It is by no means definitive and all results should be taken with a grain of salt. But they can be useful if applied properly.

Cheers,
Brian

-- 
===================================================================

Brian Peasland
dba_at_remove_spam.peasland.com

Remove the "remove_spam." from the email address to email me.


"I can give it to you cheap, quick, and good. Now pick two out of
 the three"