Depending on the size of your system, how much storage (how many DASD
units), how much memory you have, how many jobs are active in the system,
and how many new jobs start daily, these system values might be a bit low.
They all relate to the job table where there is one entry per job. So first
do WRKACTJOB and see how many active jobs there are on the system. Then do
a WRKSYSSTS to see how many jobs that reports. ( The difference in those
two is a rough approximation of how many spool files are on the system )
The initial number of jobs should be set to a number that's at least as many
as show in WRKSYSSTS.
The system will slow down marginally as new jobs are added for a very short
while (usually only milliseconds, but it's sometimes noticeable) when that
limit is reached in order to expand the job table, by the number of
additional jobs. So if your system runs a large number of jobs per day then
I'd bump the additional active and additional total up a bit. The problem
is there's no way to tell you how much to bump it so I would not go for more
than about 10% at a time until you start to see it settle down a little.
Now the warning: Memory has a significant effect on new jobs starting, jobs
ending, and file open/closes. If the system is memory constrained clearly
it will struggle to get these tasks done in order to process your jobs.
Most likely you will get much bigger effect by a proper tuning of the system
memory pools and/or adding memory than these system values will do. SQL
can add to the issue if there are not sufficient and properly constructed
index objects. So I'll bet the root cause is not these system values,
rather a couple of other things to look into.
Rob's right there will be sessions at the COMMON Annual Meeting on these
topics (I do one called Optimizing Work Management for instance) and there
are several sessions on the performance tools and performance.
--
Jim Oberholtzer
Agile Technology Architects
-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of
Hoteltravelfundotcom
Sent: Friday, April 22, 2016 7:18 AM
To: Midrange Systems Technical Discussion
Subject: system tuning question
I am looking at these:
The system seems a bit slow in response time. Not only that, but often there
is a fairly big delay in doing something simple. Running a simple command
like wrkobj at time this has taken 20 seconds to come up. Worse than this,
we have a very large Financial A/R report that when it runs full blast it
literally knocks people out of their screen on the Java XT side. We are
underpowered but till we improve that I want to see if we can tune.
QACTJOB *ALC Initial number of active jobs = 200
QADLACTJ *ALC Additional number of active jobs = 30
QADLTOTJ *ALC Additional number of total jobs = 30
If there are other system values to consider please tell me. Note, I am not
going to make changes on my own but only will present the findings or advice
from the experts here.
--
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.
Please contact support@xxxxxxxxxxxx for any subscription related questions.
midrange.com
