Re: Auto Tuning

[Glenn Ericson <Glenn-Ericson@xxxxxxx>]

Trevor,
based on your  views, would you  think it a good idea for the auto 
tuner  to save  some stats, and forecast  forward based on trend to be move 
from reactive to proactive status?

At 09:02 PM 2/2/2004 -0600, trevor perry wrote:
> Ken,
>
> I am sure there are many many theories, and the programmers of the Auto
> Tuner are the only ones who really know.
>
> My personal theory is that the Faults/Second thresholds on the WRKSHRPOOL
> are set too loose. If they were tighter, you might find the activity levels
> may go down.
> Remember, the Auto Tuner is always behind the actual "happening"
> performance, and therefore modifying future performance based on past
> resource requirements and processing. If the past work was done
> "successfully", then obviously... the AutoTuner has to put more work through
> the pool, so it increases the activity levels!!
>
> The shipped values for ALL the performance parameters tend to be very loose.
> Recently, we have spent a little effort with several customers where we
> tightened these values, resulting in more balanced performance and increased
> throughput.
>
> My 2 cents..
> Trevor
>
> ----- Original Message -----
> From: "Graap, Ken" <keg@xxxxxxxxxxxxx>
> Subject: Auto Tuning
>
>
> > I've always noticed that the Auto Tuning function of OS/400...
> >
> > .. never lowers the activity level (MAX Active from WRKSYSSTS) for a
> storage
> > pool, it just raises them slowly over time.
> >
> > Does anyone know why this is?
> >
> > Kenneth
> >
> > ****************************************
>
> _______________________________________________
> This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.