IDMS 17.0 SP1 and Above the Bar

[ca.portal.admin]

Hello All:



One of Chris's comments created a question for me. Can IDMS 17.0 SP1 use any
memory ""Above the Bar""?



William M. Allen, Jr.

ARCH Consulting Associates, Ltd.

[phone]
"
IDMS 3rd-party providers forum
[e-mail]
SMTP
[e-mail]
[e-mail]
SMTP








Normal

Normal
Re: IDMS 17.0 SP1 and Above the Bar
"I'm just a little more than two weeks out putting R17 SP1 into production.
I have a couple of observations:

1) You are going to need three LE modules for this release. You can run them
from LPA or from CDMSLIB. It doesn't matter if ZIIP=N or ZIIP=Y is
specified.

2) Running the LE modules from CDMSLIB increases above the line storage
requirements. Running from LPA has a very tiny increase in above the line
storage.

3) ZIIP = Y requires the IDMS software library to be authorized. If you
don't, IDMS complains and forces ZIIP=N.

4) DMCL processing is different. Pre 17 SP1, the DMCL is loaded into the XA
Program pool. Now it is loaded into storage pool 255. When you do a
display of storage pool 255 you'll get two entries for it. One is what is
set in the SYSGEN and the other is for the DMCL. There's a technote that
explains this.

5) You must specify ZIIP=Y to measure ZIIP potential. The enclave isn't
created until you specify ZIIP=Y. Use DCMT D SUBTASK ? to see what is going
on. Two CPU measurements are reported, time on ZIIP and ZIIP time on CP.
If you don't have a ZIIP the time will be reported as ZIIP on CP. You can
look at SMF 72 records for the same information. You won't get any ZIIP
numbers until ZIIP=Y.

If anyone running with ZIIP=Y would like to add to this list, I would
appreciate it. I've only been looking at this for 10 days and still have a
lot to learn.

Mark Grindstaff

[ca.portal.admin]

Hello All:



One of Chris's comments created a question for me. Can IDMS 17.0 SP1 use any
memory ""Above the Bar""?



William M. Allen, Jr.

ARCH Consulting Associates, Ltd.

[phone]
"
IDMS Public Discussion Forum
[e-mail]
SMTP
[e-mail]
[e-mail]
SMTP








Normal

Normal
Re: IDMS 17.0 SP1 and Above the Bar
"my comments are interspersed with ----->

Chris Hoelscher
IDMS/DB2 Database Architect
Humana Inc
502-476-2538
[e-mail]

you only need to test the programs that you want to work correctly






From:
Mark Grindstaff <[e-mail]>
To:
[e-mail]
Date:
04/29/2010 06:03 PM
Subject:
Re: [IDMSVENDOR-L] IDMS 17.0 SP1 and Above the Bar
Sent by:
IDMS 3rd-party providers forum <[e-mail]>



I'm just a little more than two weeks out putting R17 SP1 into production.
I have a couple of observations:

1) You are going to need three LE modules for this release. You can run
them
from LPA or from CDMSLIB. It doesn't matter if ZIIP=N or ZIIP=Y is
specified.

-----> yes - idms r17 REQUIRES LE runtime - i may have been the last site
in IDMSdom to run IDMS with a COBOL II runtime - but i had to change this
for R17

2) Running the LE modules from CDMSLIB increases above the line storage
requirements. Running from LPA has a very tiny increase in above the line
storage.

3) ZIIP = Y requires the IDMS software library to be authorized. If you
don't, IDMS complains and forces ZIIP=N.

-----> only the loadlib(s) containing line drivers, nuc modules, and
service drivers are loaded - you do NOT need to authorize all the
libraries in the steplib concatenation or CDMS lib concatenation


4) DMCL processing is different. Pre 17 SP1, the DMCL is loaded into the
XA
Program pool. Now it is loaded into storage pool 255. When you do a
display of storage pool 255 you'll get two entries for it. One is what is
set in the SYSGEN and the other is for the DMCL. There's a technote that
explains this.

-----> if you have an external monitor that watched for storage pool %
full - check with them - this second pool 255 will always be full and
trigger alerts unless fixes (if available) are applied


5) You must specify ZIIP=Y to measure ZIIP potential. The enclave isn't
created until you specify ZIIP=Y. Use DCMT D SUBTASK ? to see what is
going
on. Two CPU measurements are reported, time on ZIIP and ZIIP time on CP.
If you don't have a ZIIP the time will be reported as ZIIP on CP. You can
look at SMF 72 records for the same information. You won't get any ZIIP
numbers until ZIIP=Y.

-----> you may also get zIIP on CP if Ziip=Y and you have ziips and all
ziips are full for a moment (the srb will execute on CP, not wait for a
ziip to become available


If anyone running with ZIIP=Y would like to add to this list, I would
appreciate it. I've only been looking at this for 10 days and still have
a
lot to learn.

-----> not all CVs may be suited for zIIP - CVs that are heavy ADS or
COBOL will experience heavy TCB/SRB swapping - if you see this - do lots
of analysis with ZIIP on and off - you may actually be spending more
billable CPU/task WITH ziip than without - i am monitoring this now. At
my site, i have two types of CVs - one type is mostly ADS & COBOL, the
other is a CICS/LU62/UCF backend. ON the first - i average almost 1500
swaps per task - on the other, less than 1/2 swap per task - i will
eventually be turning ziip off on the first for a week or two and seeing
how that affects average cpu /task on that CV






The information transmitted is intended only for the person or entity to which it is addressed and may contain CONFIDENTIAL material. If you receive this material/information in error, please contact the sender and delete or destroy the material/information.
"
IDMS 3rd-party providers forum
[e-mail]
SMTP
[e-mail]
[e-mail]
SMTP








Normal

Normal
Re: [IDMSVENDOR-L] IDMS 17.0 SP1 and Above the Bar
"my comments are interspersed with ----->

Chris Hoelscher
IDMS/DB2 Database Architect
Humana Inc
502-476-2538
[e-mail]

you only need to test the programs that you want to work correctly






From:
Mark Grindstaff <[e-mail]>
To:
[e-mail]
Date:
04/29/2010 06:03 PM
Subject:
Re: [IDMSVENDOR-L] IDMS 17.0 SP1 and Above the Bar
Sent by:
IDMS 3rd-party providers forum <[e-mail]>



I'm just a little more than two weeks out putting R17 SP1 into production.
I have a couple of observations:

1) You are going to need three LE modules for this release. You can run
them
from LPA or from CDMSLIB. It doesn't matter if ZIIP=N or ZIIP=Y is
specified.

-----> yes - idms r17 REQUIRES LE runtime - i may have been the last site
in IDMSdom to run IDMS with a COBOL II runtime - but i had to change this
for R17

2) Running the LE modules from CDMSLIB increases above the line storage
requirements. Running from LPA has a very tiny increase in above the line
storage.

3) ZIIP = Y requires the IDMS software library to be authorized. If you
don't, IDMS complains and forces ZIIP=N.

-----> only the loadlib(s) containing line drivers, nuc modules, and
service drivers are loaded - you do NOT need to authorize all the
libraries in the steplib concatenation or CDMS lib concatenation


4) DMCL processing is different. Pre 17 SP1, the DMCL is loaded into the
XA
Program pool. Now it is loaded into storage pool 255. When you do a
display of storage pool 255 you'll get two entries for it. One is what is
set in the SYSGEN and the other is for the DMCL. There's a technote that
explains this.

-----> if you have an external monitor that watched for storage pool %
full - check with them - this second pool 255 will always be full and
trigger alerts unless fixes (if available) are applied


5) You must specify ZIIP=Y to measure ZIIP potential. The enclave isn't
created until you specify ZIIP=Y. Use DCMT D SUBTASK ? to see what is
going
on. Two CPU measurements are reported, time on ZIIP and ZIIP time on CP.
If you don't have a ZIIP the time will be reported as ZIIP on CP. You can
look at SMF 72 records for the same information. You won't get any ZIIP
numbers until ZIIP=Y.

-----> you may also get zIIP on CP if Ziip=Y and you have ziips and all
ziips are full for a moment (the srb will execute on CP, not wait for a
ziip to become available


If anyone running with ZIIP=Y would like to add to this list, I would
appreciate it. I've only been looking at this for 10 days and still have
a
lot to learn.

-----> not all CVs may be suited for zIIP - CVs that are heavy ADS or
COBOL will experience heavy TCB/SRB swapping - if you see this - do lots
of analysis with ZIIP on and off - you may actually be spending more
billable CPU/task WITH ziip than without - i am monitoring this now. At
my site, i have two types of CVs - one type is mostly ADS & COBOL, the
other is a CICS/LU62/UCF backend. ON the first - i average almost 1500
swaps per task - on the other, less than 1/2 swap per task - i will
eventually be turning ziip off on the first for a week or two and seeing
how that affects average cpu /task on that CV






The information transmitted is intended only for the person or entity to which it is addressed and may contain CONFIDENTIAL material. If you receive this material/information in error, please contact the sender and delete or destroy the material/information.
"
IDMS Public Discussion Forum
[e-mail]
SMTP
[e-mail]
[e-mail]
SMTP








Normal

Normal
Re: IDMS 17.0 SP1 and Above the Bar
"If you're at CA-World in Las Vegas - then ""above the bar memory"" is only av=
ailable as long as you are sober enough to sit or stand upright 8-) =
is one answer

""Yes"" is likely the answer you are after - cheers - Gary =20


Gary Cherlet
Justice Technology Services
Department of Justice, SA Government
Telephone [phone]
Facsimile [phone]
Mobile [phone]
MailTo:[e-mail]

This e-mail message and any attachments are qualified as follows: Addressin=
g: If you have received this e-mail in error, please advise by reply e-mai=
l to the sender. Please also destroy the original transmission and its con=
tents. Confidentiality: This e-mail may contain confidential information w=
hich also may be legally privileged. Only the intended recipient(s) may ac=
cess, use, distribute or copy this e-mail. Individual Views: Unless otherw=
ise indicated, the views expressed are those of the sender, not Justice Tec=
hnology Services. Computer Viruses: It is the recipient's responsibility t=
o check the e-mail and any attached files for viruses.