Re: Query

[rob@xxxxxxxxx]

There are several ways to limit how much of your total CPU is used by any 
one Query.

One is to use multiple processors and NOT install SNMP.  For example, on 
our 840 12way the most any one Query user may use is roughly 8% or 1 cpu.

Perhaps there are also some options you can use in the QAQQINI, or Query 
options, file.  Searching the infocenter for QAQQINI might help.

Another method is to throw off your Query's to your lpar, or a second 
machine.  If you are already duplicating your data due to High 
Availability software this can be attractive.

You can flat out restrict Query from running in interactive by removing 
*INTERACT from CHGCMD CMD(RUNQRY) ALLOW(*IMOD *BMOD *IREXX *BREXX *BPGM 
*IPGM *EXEC *INTERACT *BATCH)

You could also look at options, like running the Query in a different 
subsystem.

Another thing to consider, for those disk filling runaway queries is to 
change the user, or group, profile to only allow so much disk in CHGUSRPRF 
... MAXSTG(####)  We keep our payroll and accounting group profile 
hovering around 95%.  And, actually I've found that I only have to kick 
that number up around once a year.  It's pretty static.  (Does PRTRAND 
sound familiar to you?)  But it did stop the one user (a developer with 
*ALLOBJ) from running a query that actually filled up our disk (which is 
currently around 75%).

Rob Berendt
-- 
Group Dekko Services, LLC
Dept 01.073
PO Box 2000
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





<SRamanujan@xxxxxxxxxxxxxxxxxx> 
Sent by: midrange-l-bounces@xxxxxxxxxxxx
08/17/2004 12:39 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
<midrange-l@xxxxxxxxxxxx>
cc

Fax to

Subject
Query






Hi all,

I do not know much about LPAR and IBM site is down on this topic.

I am interested to know how I can benchmark a whole set of processes in
terms of CPU used and not the time involved in the process.  I am not
sure if I am asking the right question here.  Since I see the number of
minutes the process took but not how much CPU was utilized in many of
the bench marks.  Is there a ratio for the CPU used with time?  Is LPAR
the best way to do a benchmark?  How can I set this up? 

Also, I would like to restrict the user to not use more than certain
threshold of CPU, is this possible?  Or is the job priority & subsystem
the solution for this or LPAR?  Can anyone please send me any material
on LPAR? 

Thanks,
Sudha


Sudha Ramanujan
SunGard Futures Systems
sramanujan@xxxxxxxxxxxxxxxxxx
(312) 577 6179
(312) 577 6101 - Fax


--
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.