Re: Balancing disk drives

I DID use the "add and balance".  And it automatically did it twice.
Job log from job that started it:
Job 514520/ROB/DSP01 started on 07/26/11 at 14:03:53
 QSYS. Job entered system on 07/26/11 at 14:03:53.
Message queue ROB is allocated to another job.
STRSST
wrkdsksts
signoff
Job 514520/ROB/DSP01 ended on 07/26/11 at 15:54:58;

DSPLOG
07/26/11-15:50:23: ASP balancing successfully started for ASP 1.
07/26/11-19:55:55: ASP balancing for ASP 1 ended.
07/26/11-19:55:55:  ASP balancing successfully started for ASP 1.
07/26/11-20:23:38: ASP balancing for ASP 1 ended.

The ONLY other interactive type job during that time was Job
514488/PEGGY/GDSL621.  While Peggy has that authority and is a great
operator when it comes to following a script she has a small comfort zone
and she would NEVER run STRSST or STRASPBAL and hasn't written CL in over
a decade.  She get's on there just to get her BRMS stuff and that's it.

Job load is tough to measure.  This a Mimix backup lpar for another lpar
on a different box in a different city.  The same jobs are all that runs
"Did we get any journal entries, if so, post them".  Reads are at a
minimum.  No interactive, no batch.  Where I really noticed the
performance was in the nightly BRMS backup.  Added HOURS to the run.
Granted, the first night it was still balancing the disk drive.  Backup
started at 17:30 and the balance didn't finish until 20:23.  Following
night was still about 1.5 hours longer than prior to adding the disk arm.
After the *USAGE balance we're down to only an additional 20 minutes.


Rob Berendt
--
Group Dekko
Dept 1600
Mail to:  2505 Dekko Drive
         Garrett, IN 46738
Ship to:  Dock 108
         6928N 400E
         Kendallville, IN 46755
http://www.dekko.com





From:   DrFranken <midrange@xxxxxxxxxxxx>
To:     midrange-l@xxxxxxxxxxxx
Date:   07/29/2011 09:21 AM
Subject:        Re: Balancing disk drives
Sent by:        midrange-l-bounces@xxxxxxxxxxxx



OK But it sounds as if you made a tactical error when you first added it
by not using the "Add AND Balance" option. Rather it sounds like you did
an "Add" followed by a "Balance" (with *CAPACITY).   These sound very
very close but they are very much NOT.   The key difference is that with
the former the drive is flagged to prevent any data getting on it until
the balance is completed. Thus the system will FIRST fill it with an
'average load' of stuff from the other drives and THEN make it available
for new data. The latter process omits that flag and thus all the newest
and frequently hottest data goes to that drive almost continuously until
it's about evenly full with the other drives. The new drive in this case
goes "hot".

Then you undertook the process of rebalancing for better performance.
You collected three hours of data but in that three hours did you
properly use all the stuff on that new drive? There is no guarantee that
in that time frame you succeeded in running the right jobs to enable the
system to perfectly balance that drive.

Finally while, yes, 10 is five times as much as 2, it's still only 10
which ain't very busy.  Over time that will only get better.

    - Larry "DrFranken" Bolhuis

On 7/29/2011 8:50 AM, rob@xxxxxxxxx wrote:

*This message was transferred with a trial version of CommuniGate(r)

Pro*

It was not well balanced on a usage standpoint previously.  We just

added

the drive this week and it dumped all new stuff into it.  It only took 3
hours to balance it on usage.  Not counting the hours spent figuring the
usage by tracing.

What galled me was that the drive continues to get hammered.  Yes you're
right, the first snapshot with the 15/100 ratio was quick.  However
Elapsed time:   00:35:14
   %
Busy
...
    2
    2
    2
    1
    2
    2
   10
still shows it out of whack.


Rob Berendt

--
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.


--
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.