Re: ZIIP engine utilization

[Richards, Robert B.]

Martin, 

Can I get a copy of your presentation?

Bob

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Martin Packer
Sent: Friday, May 18, 2018 4:01 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZIIP engine utilization



I have a zIIP Capacity Planning presentation - which I last gave yesterday

in London at Z Technical University in London. It might well be of use.



Hint: You don’t want to go anywhere near 100% zIIP Pool Utilisation.



Cheers, Martin



Sent from my iPad



> On 18 May 2018, at 14:37, Scott Chapman <scott.chap...@epstrategies.com>

wrote:

>

> Since you have at least the RMF post processor reports available to you,

the CPU Activity report is a high-level place to start. In particular I

would start at the Partition Data Report section. Look for the total

physical utilization of the zIIP(s).

>

> That's not the only place you may see an indication of a problem. I.E.

your zIIP (or your GPs for that matter) may be much less than 100% busy and

you may still be having performance problems. There's lots of things to

consider.

>

> One example is that the RMF interval length can hide spikes that may be

causing problems. E.G. the 15 minute average utilization may be 40% busy,

but that may be because it's 100% busy for a solid 5 minutes, and mostly

unused for the remainder of the interval. Looking at the RMF III panels

interactively can be useful to help see this. (And there's some other very

useful SMF records as well.)

>

> Or maybe the LPAR weights are not correct for the workload and your

business needs.

>

> Or maybe the WLM policy is not protecting/helping the work that really

needs to be protected/helped.

>

> Or maybe it's not so much a zIIP capacity problem as it is a GP capacity

issue.

>

> Regardless, we're always happy to do a free cursory performance review.

We often find "interesting" things during those reviews. See:

https://www.pivotor.com/cursoryReview.html



>

> Scott Chapman

>

> --

> For IBM-MAIN subscribe / signoff / archive access instructions,

> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

>Unless stated otherwise above:

IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 

Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU




--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Martin Packer]

I have a zIIP Capacity Planning presentation - which I last gave yesterday
in London at Z Technical University in London. It might well be of use.

Hint: You don’t want to go anywhere near 100% zIIP Pool Utilisation.

Cheers, Martin

Sent from my iPad

> On 18 May 2018, at 14:37, Scott Chapman 
wrote:
>
> Since you have at least the RMF post processor reports available to you,
the CPU Activity report is a high-level place to start. In particular I
would start at the Partition Data Report section. Look for the total
physical utilization of the zIIP(s).
>
> That's not the only place you may see an indication of a problem. I.E.
your zIIP (or your GPs for that matter) may be much less than 100% busy and
you may still be having performance problems. There's lots of things to
consider.
>
> One example is that the RMF interval length can hide spikes that may be
causing problems. E.G. the 15 minute average utilization may be 40% busy,
but that may be because it's 100% busy for a solid 5 minutes, and mostly
unused for the remainder of the interval. Looking at the RMF III panels
interactively can be useful to help see this. (And there's some other very
useful SMF records as well.)
>
> Or maybe the LPAR weights are not correct for the workload and your
business needs.
>
> Or maybe the WLM policy is not protecting/helping the work that really
needs to be protected/helped.
>
> Or maybe it's not so much a zIIP capacity problem as it is a GP capacity
issue.
>
> Regardless, we're always happy to do a free cursory performance review.
We often find "interesting" things during those reviews. See:
https://www.pivotor.com/cursoryReview.html

>
> Scott Chapman
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Scott Chapman]

Since you have at least the RMF post processor reports available to you, the 
CPU Activity report is a high-level place to start. In particular I would start 
at the Partition Data Report section. Look for the total physical utilization 
of the zIIP(s). 

That's not the only place you may see an indication of a problem. I.E. your 
zIIP (or your GPs for that matter) may be much less than 100% busy and you may 
still be having performance problems. There's lots of things to consider. 

One example is that the RMF interval length can hide spikes that may be causing 
problems. E.G. the 15 minute average utilization may be 40% busy, but that may 
be because it's 100% busy for a solid 5 minutes, and mostly unused for the 
remainder of the interval. Looking at the RMF III panels interactively can be 
useful to help see this. (And there's some other very useful SMF records as 
well.)

Or maybe the LPAR weights are not correct for the workload and your business 
needs. 

Or maybe the WLM policy is not protecting/helping the work that really needs to 
be protected/helped. 

Or maybe it's not so much a zIIP capacity problem as it is a GP capacity issue. 

Regardless, we're always happy to do a free cursory performance review. We 
often find "interesting" things during those reviews. See: 
https://www.pivotor.com/cursoryReview.html 

Scott Chapman

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Christopher Y. Blaicher]

SMF30_TIME_zIIP_ON_CP is another indicator.  Look at subtype 2 interval records 
for long running tasks such as CICS and DB2 and at subtype 4 records for 
regular jobs.

SMF30_TIME_zIIP_ON_CP is all the work that could be done on a z/IIP but 
couldn't due to the ZIIP was busy with other work.

Also, there is a parm in IEAOPTxx called ZIIPAWMT that is the number of 
microseconds that ZIIP eligible work might wait to be dispatched on a ZIIP 
before the work is dispatched on a GP.  The theory is that it will wait a 
period of time to see if a ZIIP becomes available to run the work.  The problem 
is while the work is waiting, it isn't doing anything.  This is why you don't 
try to run ZIIPs at anything close to 100%.

Chris Blaicher
Technical Architect
Mainframe Development
P: 201-930-8234  |  M: 512-627-3803
E: cblaic...@syncsort.com

Syncsort Incorporated
2 Blue Hill Plaza #1563
Pearl River, NY 10965
www.syncsort.com

Data quality leader Trillium Software is now a part of Syncsort.

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Richards, Robert B.
Sent: Friday, May 18, 2018 6:46 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZIIP engine utilization

IIRC, you can look at zIIP-eligible time (CPUZIETM). If it is a high value, you 
are not only losing the benefit of running on an additional uncapped specialty 
processor, you are potentially wasting money if then running on a GP at a time 
that drives the system to a new 4HRA on software costs!

I justified the purchase of zIIP solely on this basis when I was able to 
demonstrate a ROI in five months due to eliminating the potential increase in 
costs attributed to running on a GP during the setting of a new peak four hour 
rolling average. YMMV

Props to Al Sherkow and his LCS software to be able to prove my assertions.

Bob

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Vernooij, Kees (ITOPT1) - KLM
Sent: Friday, May 18, 2018 2:36 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZIIP engine utilization

The situation for a zIIP is somewhat more complex and 2-staged.
If a zIIP becomes heavier used, delays for the zIIP begin to show up. It is at 
30% utilization when you have 1 zIIP, 60% utilization when you have 2 zIIP.
It the zIIP becomes heavier loaded and delays are growing, the CP is asked to 
assist the zIIP. The is also reported in RMF.

Kees.


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU]
> On Behalf Of Barbara Nitz
> Sent: 18 May, 2018 7:57
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: ZIIP engine utilization
>
> > Can you point me to the best place to see if Ziip is getting maxed
> out.
>
> Look at the TYPE70 SMF records. They will show you if the ZIIP is 100%
> busy.
>
> Regards, Barbara
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions, send
> email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

For information, services and offers, please visit our web site: 
http://www.klm.com. This e-mail and any attachment may contain confidential and 
privileged material intended for the addressee only. If you are not the 
addressee, you are notified that no part of the e-mail or any attachment may be 
disclosed, copied or distributed, and that any other action related to this 
e-mail or attachment is strictly prohibited, and may be unlawful. If you have 
received this e-mail by error, please notify the sender immediately by return 
e-mail, and delete this message.

Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its 
employees shall not be liable for the incorrect or incomplete transmission of 
this e-mail or any attachments, nor responsible for any delay in receipt.
Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch 
Airlines) is registered in Amstelveen, The Netherlands, with registered number 
33014286



--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN





ATTENTION: -

The information contained in this message (including any files transmitted with 
this message) may contain proprietary, trade secret or other confidential 
and/or legally privileged information. Any pricing information contained in 
this message or in any files transmitted with this message is always 
confide

Re: ZIIP engine utilization

[Seymour J Metz]

I agree with zman; you need to get some performance expertise. If you have 
slack time and management is willing to pay for training, that's a good option. 
Otherwise they need to hire someone with performance and tuning expertice.

It's not rocket science, but you will need time to learn and overtime should be 
only for emergencies, not a way of life.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Discussion List  on behalf of 
Jesse Lynch 
Sent: Thursday, May 17, 2018 7:40 PM
To: IBM-MAIN@listserv.ua.edu
Subject: ZIIP engine utilization

Our senior performance person has retired.  We have a ZIIP on a Z13s.  One of 
our customers is using Adabase and some
product to make maximum use of the Ziip.   They feel that the Ziip has maxed 
out from I think messages they are getting from their product and response is 
slower.  I am getting this third hand so don't know a lot of details.  Can you 
point me
to the best place to see if Ziip is getting maxed out.   I am looking at the 
RMF Batch CPU report.  I see some % usages
for the IIP.  Also see that one Lpar (not even sure which Lpar yet they are 
talking about) seems to say its WLM capped.
I have looked at the Lpar definitions on the HMC and nothing jumps out at me.

If you have any comments I would appreciate it.  Thanks.  Jess

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Richards, Robert B.]

IIRC, you can look at zIIP-eligible time (CPUZIETM). If it is a high value, you 
are not only losing the benefit of running on an additional uncapped specialty 
processor, you are potentially wasting money if then running on a GP at a time 
that drives the system to a new 4HRA on software costs! 

I justified the purchase of zIIP solely on this basis when I was able to 
demonstrate a ROI in five months due to eliminating the potential increase in 
costs attributed to running on a GP during the setting of a new peak four hour 
rolling average. YMMV

Props to Al Sherkow and his LCS software to be able to prove my assertions.

Bob

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Vernooij, Kees (ITOPT1) - KLM
Sent: Friday, May 18, 2018 2:36 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZIIP engine utilization

The situation for a zIIP is somewhat more complex and 2-staged.
If a zIIP becomes heavier used, delays for the zIIP begin to show up. It is at 
30% utilization when you have 1 zIIP, 60% utilization when you have 2 zIIP.
It the zIIP becomes heavier loaded and delays are growing, the CP is asked to 
assist the zIIP. The is also reported in RMF.

Kees.


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Barbara Nitz
> Sent: 18 May, 2018 7:57
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: ZIIP engine utilization
> 
> > Can you point me to the best place to see if Ziip is getting maxed
> out.
> 
> Look at the TYPE70 SMF records. They will show you if the ZIIP is 100%
> busy.
> 
> Regards, Barbara
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

For information, services and offers, please visit our web site: 
http://www.klm.com. This e-mail and any attachment may contain confidential and 
privileged material intended for the addressee only. If you are not the 
addressee, you are notified that no part of the e-mail or any attachment may be 
disclosed, copied or distributed, and that any other action related to this 
e-mail or attachment is strictly prohibited, and may be unlawful. If you have 
received this e-mail by error, please notify the sender immediately by return 
e-mail, and delete this message. 

Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its 
employees shall not be liable for the incorrect or incomplete transmission of 
this e-mail or any attachments, nor responsible for any delay in receipt. 
Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch 
Airlines) is registered in Amstelveen, The Netherlands, with registered number 
33014286



--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Vernooij, Kees (ITOPT1) - KLM]

The situation for a zIIP is somewhat more complex and 2-staged.
If a zIIP becomes heavier used, delays for the zIIP begin to show up. It is at 
30% utilization when you have 1 zIIP, 60% utilization when you have 2 zIIP.
It the zIIP becomes heavier loaded and delays are growing, the CP is asked to 
assist the zIIP. The is also reported in RMF.

Kees.


> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Barbara Nitz
> Sent: 18 May, 2018 7:57
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: ZIIP engine utilization
> 
> > Can you point me to the best place to see if Ziip is getting maxed
> out.
> 
> Look at the TYPE70 SMF records. They will show you if the ZIIP is 100%
> busy.
> 
> Regards, Barbara
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

For information, services and offers, please visit our web site: 
http://www.klm.com. This e-mail and any attachment may contain confidential and 
privileged material intended for the addressee only. If you are not the 
addressee, you are notified that no part of the e-mail or any attachment may be 
disclosed, copied or distributed, and that any other action related to this 
e-mail or attachment is strictly prohibited, and may be unlawful. If you have 
received this e-mail by error, please notify the sender immediately by return 
e-mail, and delete this message. 

Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its 
employees shall not be liable for the incorrect or incomplete transmission of 
this e-mail or any attachments, nor responsible for any delay in receipt. 
Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch 
Airlines) is registered in Amstelveen, The Netherlands, with registered number 
33014286



--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Barbara Nitz]

> Can you point me to the best place to see if Ziip is getting maxed out. 

Look at the TYPE70 SMF records. They will show you if the ZIIP is 100% busy.

Regards, Barbara

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[Mike Hochee]

I'm sorry, but the following will not answer your immediate question about how 
to determine if a zIIP is getting maxed out. I don't know the answer to your 
specific question. These comments pertain more to a longer view of performance 
monitoring.   

In the interim though, you can always try to answer an important question when 
it comes to fighting performance related fires - What changed?  

Almost all of the fires I've worked on at various shops, some with many SLAs in 
place, were directly related to application/system changes where the 
anticipated performance impact was either underestimated, not well understood, 
or never evaluated. If your shop is big on change management that includes; 
applications, database, sysprogs, security, network, capacity planning and 
workload management, etc. then you may have available a decent record of what 
actually changed, when they took effect, and eventually how those changes 
relate to different workloads on your system.  

What kind of performance monitoring/reporting tools are available?  You may 
want to get extremely familiar with some of the daily reports and resource 
utilization values for critical workloads. The idea is to get to know your 
system, first from 50,000', then from 10,000', then 2,000' etc., until you get 
granular enough to be pointing your finger at a Natural program that has gone 
off the rails, or pig CICS transaction, or whatever. 

Until you know the performance characteristics of each workload on the system, 
how they relate, and what's changed recently, it would be pretty tough to make 
performance and tuning recommendations (of any kind), and expect positive 
results.  

Identifying and cultivating useful working relationships with SME-type 
individuals in different departments can also go a long way, then you will have 
more/less confidence in what you're hearing in the midst of some frenzied 
performance event. 

Regarding Adabas, three simple things to keep an eye on are buffer pool 
efficiency, format translations and format overwrites. These are useful CPU 
related metrics because they are usually actionable. DBAs can make adjustments 
and/or have application developers make changes to reduce CPU utilization (with 
GP and potentially zIIP impacts).  Your DBAs can execute very low overhead 
dstat commands to track these metrics at whatever interval is appropriate. 

If you aren't already familiar with them, ABCs of Systems Programming Vol 11 is 
good, as well as the z/OS RMF User's Guide and Report Analysis manuals.  

HTH, 
Mike 
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of zMan
Sent: Thursday, May 17, 2018 8:39 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZIIP engine utilization

Not trying to be a jerk, but you're playing with fire here. Call Cheryl Watson 
and pay for some help. Take some classes. Go to SHARE and learn. If management 
is too cheap to pay for the expertise, update your resume and go find a real 
job.

I'd explain it to them this way: If the lead engineer at one of the nuclear 
power plants retired and nobody knew how to do what he'd been doing, would they 
suggest asking for advice on a list?

Not that folks here will deliberately lead you astray, of course. But 
performance work is too highly skilled to fake your way through it: that's just 
asking for a disaster.

On Thu, May 17, 2018 at 7:40 PM, Jesse Lynch <jesse.ly...@state.mn.us>
wrote:

> Our senior performance person has retired.  We have a ZIIP on a Z13s.  
> One of our customers is using Adabase and some
> product to make maximum use of the Ziip.   They feel that the Ziip has
> maxed out from I think messages they are getting from their product 
> and response is slower.  I am getting this third hand so don't know a 
> lot of details.  Can you point me
> to the best place to see if Ziip is getting maxed out.   I am looking at
> the RMF Batch CPU report.  I see some % usages for the IIP.  Also see 
> that one Lpar (not even sure which Lpar yet they are talking about) 
> seems to say its WLM capped.
> I have looked at the Lpar definitions on the HMC and nothing jumps out 
> at me.
>
> If you have any comments I would appreciate it.  Thanks.  Jess
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions, send 
> email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>



--
zMan -- "I've got a mainframe and I'm not afraid to use it"

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: ZIIP engine utilization

[zMan]

Not trying to be a jerk, but you're playing with fire here. Call Cheryl
Watson and pay for some help. Take some classes. Go to SHARE and learn. If
management is too cheap to pay for the expertise, update your resume and go
find a real job.

I'd explain it to them this way: If the lead engineer at one of the nuclear
power plants retired and nobody knew how to do what he'd been doing, would
they suggest asking for advice on a list?

Not that folks here will deliberately lead you astray, of course. But
performance work is too highly skilled to fake your way through it: that's
just asking for a disaster.

On Thu, May 17, 2018 at 7:40 PM, Jesse Lynch 
wrote:

> Our senior performance person has retired.  We have a ZIIP on a Z13s.  One
> of our customers is using Adabase and some
> product to make maximum use of the Ziip.   They feel that the Ziip has
> maxed out from I think messages they are getting from their product and
> response is slower.  I am getting this third hand so don't know a lot of
> details.  Can you point me
> to the best place to see if Ziip is getting maxed out.   I am looking at
> the RMF Batch CPU report.  I see some % usages
> for the IIP.  Also see that one Lpar (not even sure which Lpar yet they
> are talking about) seems to say its WLM capped.
> I have looked at the Lpar definitions on the HMC and nothing jumps out at
> me.
>
> If you have any comments I would appreciate it.  Thanks.  Jess
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>



-- 
zMan -- "I've got a mainframe and I'm not afraid to use it"

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN