Re: Is Oracle Simply a Pig - The Worst I've See

In article <MLOENNRO.93Dec1114426_at_demo1.se.oracle.com>, mloennro_at_se.oracle.com (Magnus Lonnroth) writes:
> Jim> Jim Kramer Motorola
>
> The original poster was complaining about excessive memory demands
> by Oracle front-end processes - not I/O contention. Oracle should
> never be I/O bound, it should always be CPU bound - that's our whole
> strategy. CPUs get cheaper and faster all the time, disks don't (at
> least not yet).
>
> To the original poster: get in touch with your local Oracle office and
> get some assistance in tuning memory usage. Or (I know you'll hate me
> for this) buy more memory.
>
> --
>
> Magnus Lonnroth
> Tech.Sales & Consultant
> Oracle Sweden
> Mail: mloennro_at_oracle.com

Actually, my concern was the amount of CPU required for the tiny amounts of I/O. Some of the question's posted by our DB group involved questions that involved memory.

But, just a few Oracle users ate up the CPU of a 6630. A few Oracle report writers ate up a 4000-400.

And, yes, we have had Oracle consultants tell us our system is well tuned, the DB is well tuned, and we've had tons of memory thrown at the beast. Our SGA's are huge. Our working sets incredible. that was all documented in the original note.

But, using the tools listed in the original note, to accomplish so little, we get huge compute queues. Please go back and read the original note. We have bought more memory.

Bob Comarow
Grumman Melbourne System's Division

posting from comarow_at_Eisner.decus.org

sometimes mail Received on Wed Dec 01 1993 - 14:44:10 CET