Re: wrksyssts non DB-fault paging

["Steve Landess" <sjl_123@xxxxxxxxxxx>]

"Jerry" <>
> Subject: wrksyssts non DB-fault paging
> Here are some snapshots of wrksyssts on our 803 over a period
> of about one minute.
> My concern is with the Non-DB Fault column which shows the
>system paging out to the disk...perhaps unncessarily.

> The question is how to remedy this situation.  Is more memory required?
> We have 8GB.

Jerry,

AFAIK, One minute is not an adequate time interval for measurement of
performance data on the WRKSYSSTS screen.

Five minutes is closer to the the recommended interval, IIRC.
After this amount of time has elapsed, you have a more accurate measurement.

However, off the top of my head,

1) Why do you have over 45,000 jobs in the system?
You're probably keeping too many spooled files around...
Remember, the system has to keep up with these jobs even after they have
ended.

2) Also, press F11 on this screen and also take a look at the
Wait -> Inel and Active -> Inel statistics.

3) You probably don't have enough memory dedicated to the *BASE memory pool.
My rule of thumb is to start out with around 20% of total memory allocated
to *BASE.

AND - You should NEVER have Wait -> Inel or Active -> Inel transitions
occurring in the *BASE memory pool.  If you do, you don't have enough memory
there!

4) How many interactive users do you have?  I'm assuming that pool 2 is the
interactive memory pool. I would possibly look at reducing the number of
activity levels...

5) I'm assuming pool 3 is *SPOOL...it looks OK.

6) I'm assuming that pool 4 is for batch jobs.
How many batch jobs do you allow to execute simulaneously?
I'm thinking that 159 activity levels is TOO many...

Steve Landess
Austin, Texas
(512) 423-0935