IFAHONORPRIORITY (Was: Re: zAAP Eligible Work)

2007-04-13 Thread John Eells
Below, Tom posted the book's description of the original 
implementation of IFAHONORPRIORITY.  But customers told us that 
they would much prefer that we filled up the zAAPs before using 
CP resource for Java workload.  So in z/OS R8, and also with 
APARs OA14131 and OA13953 for z/OS R6 and R7, we changed it to do 
that (more or less):


*YES*

Specifies that standard processors run both zAAP processor 
eligible and non-zAAP processor eligible work in priority order 
when the zAAP processors indicate the need for help from standard 
processors. The need for help is determined by the alternate wait 
management (AWM) function of SRM for both standard and zAAP 
processors. Standard processors help each other and standard 
processors can also help zAAP processors if YES is in effect. 
Specifying yes does not mean the priorities will always be 
honored because the system manages dispatching priorities based 
on the goals provided in the WLM service definition. AWM should 
not be disabled when IFAHONORPRIORITY=YES is in effect. See the 
description for parameter CCCAWMT for a description of AWM.


If zAAP processors are defined to the LPAR but are not online, 
the zAAP processor eligible work units are processed by standard 
processors in priority order. The system ignores the 
IFAHONORPRIORITY parameter in this case and handles the work as 
if it had no eligibility to zAAP processors. The zAAP processor 
eligible processor times are reported in RMF and SMF for planning 
purposes. Standard processors can also run zAAP processor 
eligible work (even if IFAHONORPRIORITY is set to NO), if 
necessary to resolve contention for resources with non-zAAP 
processor eligible work.


IBM suggests that you specify or default to IFAHONORPRIORITY=YES.


*NO*

Specifies that standard processors will not examine zAAP 
processor eligible work regardless of the demand for zAAP 
processors as long as there is standard processor eligible work 
available.


Tom Moulder wrote:

Something else to consider are two IEAOPTxx parameters for WLM control of
zAAP eligible workloads.

IFACROSSOVER controls whether standard CPs will be used for zAAP eligible
work.  If YES (the default), then standard CPs will be used along with zAAP
CPs.  zAAP CPs will be preferred, but standard CPs are not excluded from
consideration based on workload.  If NO, then standard CPs are not
considered for zAAP eligible work unless there are no zAAP CPs present.

IFAHONORPRIORITY controls the workload priority settings for zAAP eligible
work that is executed on standard CPs.  If YES (the default), then all work
on a standard CP will execute based on WLM assigned priority.  If NO, then
all zAAP eligible work that is assigned for execution on a standard CP will
be scheduled at a lower priority than non-zAAP eligible workloads.


snip

--
John Eells
z/OS Technical Marketing
IBM Poughkeepsie
[EMAIL PROTECTED]

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: IFAHONORPRIORITY (Was: Re: zAAP Eligible Work)

2007-04-13 Thread Tom Moulder
Thank you John, I was not aware of those changes.

Tom Moulder


-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of John Eells
Sent: Friday, April 13, 2007 7:59 AM
To: [EMAIL PROTECTED]
Subject: IFAHONORPRIORITY (Was: Re: zAAP Eligible Work)

Below, Tom posted the book's description of the original 
implementation of IFAHONORPRIORITY.  But customers told us that 
they would much prefer that we filled up the zAAPs before using 
CP resource for Java workload.  So in z/OS R8, and also with 
APARs OA14131 and OA13953 for z/OS R6 and R7, we changed it to do 
that (more or less):

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Chase, John
 -Original Message-
 From: IBM Mainframe Discussion List On Behalf Of Jim Marshall
 
 Went off to one of my ISV vendors and asked if the JAVA 
 coding in their product would be run on the zAAP processor I 
 am implementing on my two z9BC machines.  The response was 
 
 In order for XXX to use the ZAAP engines, it would have 
 to be started in a special enclave eligible for ZAAP 
 processing.  We don't do that.
 
 Does this ring a bell with anyone as being a true statement.

Sounds like it got confused with how work becomes eligible for running
on a zIIP (SRB enclave).  AFAIK (which ain't very far), the z/OS JVM
decides what runs on the zAAP, and Java code runs in a JVM.

-jc-

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Craddock, Chris
 Are you aware of what other work might be zAAP eligible?
 John, publicly, no.  The only workload that I know of today is Java.

Publicly or privately the answer is still no. zAAP processors run JAVA
code in the IBM-supplied JVM. Period. IBM has disclosed statements of
direction concerning certain functions that may become eligible to run
on the offload engines in a future release but as far as I know zAAPs
will still be JAVA only. 

As Ed mentioned somewhere earlier, the interface to the zAAP has not
been disclosed - although a few people outside IBM have a pretty good
idea how it works. For all practical purposes it is a private interface
between the JVM and the dispatcher and even though we know how it works,
it would be a direct violation of the TCs to even try to circumvent it.
In contrast, the interface to the zIIP -is- disclosed to vendors under
license. Meaning we know but we can't tell you. Both are intentionally
limited in the work they can run.

 it's touted as a Java processor, 'cuz that's the only thing using it
 today.  Read between the lines from this redbook:

Don't bother reading between the lines. It's touted as a JAVA processor
because it is a JAVA processor. It is just an ordinary z cpu engine, but
the OS makes sure that only JVM work runs on it. Ergo it's a JAVA
processor.

 If it were only destined to be a java processor, sans any copyright
 considerations or politically correct names, I would have expected a
name
 like zJAP (LOL).  Now, on the other hand, they hit the nail on the
head
 when they named the IFL.
 
 It's an application assist processor, or formally, integrated
facility
 for applications - read between the lines - reduce mainframe TCO -
don't
 charge for software - more open source leverage...  blah, blah, blah.

The zAAP -name- came very late in the game. When it was first disclosed
to us it was called an IFA, which probably stood for integrated
facility for applications as you suggest. However, after the marketers
had their way with it, the name became zAAP. That happens a lot. The
lesson is not to read any intelligent meaning into the names. Often
enough the names are just the product of a fevered marketing brain. 

CC

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread John Eells

Craddock, Chris wrote:
snip

The zAAP -name- came very late in the game. When it was first disclosed
to us it was called an IFA, which probably stood for integrated
facility for applications as you suggest. However, after the marketers
had their way with it, the name became zAAP. That happens a lot. The
lesson is not to read any intelligent meaning into the names. Often
enough the names are just the product of a fevered marketing brain. 

snip

...then again, though I've no idea what happened in this case 
(despite my job title and my perhaps sometimes-fevered brain, I'm 
not actually in Marketing), sometimes the original name does not 
make it through the name search and it gets changed for that reason.


And IIRC, IFA did indeed originally stand for what you suggested 
above.


--
John Eells
z/OS Technical Marketing
IBM Poughkeepsie
[EMAIL PROTECTED]

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Edward Jaffe

Todd Burch wrote:

If it were only destined to be a java processor, sans any copyright
considerations or politically correct names, I would have expected a name
like zJAP (LOL).  Now, on the other hand, they hit the nail on the head
when they named the IFL.
  


The original name was 'JAF'. I made a joke during the disclosure that, 
when IBM would eventually decide to extend the technology with 
'Functional Extensions' (JAFFE), I would not claim infringement of any sort.


By the time the hardware and software was developed/tested, it had been 
renamed to 'IFA'. That name persists today.


The official name was changed to 'zAAP' by marketing people at 
(practically) the last minute. Far too late to make any sort of change 
in already-developed hardware and software.


--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread David Andrews
Todd Burch wrote:
 If it were only destined to be a java processor, sans any copyright
 considerations

Since Java is trademarked by Sun, that's probably all it takes to
furrow the brows of IBM lawyers.  Hence no zJAP (or JAFFE, or anything
that contains a J).

-- 
David Andrews
A. Duda and Sons, Inc.
[EMAIL PROTECTED]

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Barry Merrill
In my opinion, zAAPs, and zIIPS and zFLS and all of the newfangled
processor type were created solely by IBM hardware gurus who could
not get IBM Marketing to reduce the prices of their CPs and the
software running on their CPs, and these hardware types knew that
if marketing wasn't going to reduce the price of z/SYSTEM hardware,
that they'd have no jobs in the future, and they invented these new
engines, and all of the software scheduling, monitoring, managing,
etc. so they could keep their jobs creating z/SYSTEM hardware. 

And of course, now, we users have to learn how to introduce these new 
engines, how to set parameters (e.g., this thread), how to select what
work can be used on them, how to recover and account their costs, i.e., we have 
to spend our time (our money) on our people costs,
which means our companies have to pay a lot more people money to save any 
hardware money on these cheaper engines.

I can see NO virtue in these processors as technical designs, as the 
added software decision logic to decide on what to run where is unquestionably 
an unnecessary overhead, and technically we'd be far
better served, with less overhead and no new people training costs, had IBM 
marketing instead reduced the prices of their CPs and
associated CP-driven software costs.

That the hardware guys had to invent these devices so they could be
priced cheaper - now those guys do deserve great credit for solving
a problem they shouldn't have had to solve - and for making it possible
for future zSYSTEMS to exist in the artificial pricing world of IBM.

Barry Merrill

Herbert W. Barry Merrill, PhD
President-Programmer
Merrill Consultants
MXG Software 
10717 Cromwell Drive
Dallas, TX 75229
www.mxg.com 


P.S. My opinion has been affirmed by many leading IBM Technicians,
 privately.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Craddock, Chris
Barry wrote
 In my opinion, zAAPs, and zIIPS and zFLS and all of the newfangled
 processor type were created solely by IBM hardware gurus who could
 not get IBM Marketing to reduce the prices of their CPs and the
 software running on their CPs, and these hardware types knew that
 if marketing wasn't going to reduce the price of z/SYSTEM hardware,
 that they'd have no jobs in the future, and they invented these new
 engines, and all of the software scheduling, monitoring, managing,
 etc. so they could keep their jobs creating z/SYSTEM hardware.
snip technically we'd be far
 better served, with less overhead and no new people training costs,
had
 IBM marketing instead reduced the prices of their CPs and
 associated CP-driven software costs.
 
 That the hardware guys had to invent these devices so they could be
 priced cheaper - now those guys do deserve great credit for solving
 a problem they shouldn't have had to solve - and for making it
possible
 for future zSYSTEMS to exist in the artificial pricing world of IBM.

Weell... not quite. Your're dead right on all counts with respect to
the costs and the overheads (naturally) but the idea didn't just spring
fully formed from the hardware guys and it wasn't directly in response
to Neanderthal marketing guys either. There was clearly a desperate need
for a new pricing model and the simplest solution would just be to have
different (SMF) accounting counters depending on what work was being
done. 

However, that idea would not pass muster with the LAWYERS - who said, in
effect, you can't charge two different prices for the same thing
without getting us flayed by the DOJ. Leaving aside for a moment the
question of whether that was really true in a legal sense (IBM doesn't
seem to have any trouble at all in charging different prices for the
same thing, to different customers all over the US and all over the
world) there had to be a solution that involved a physically different
configuration to attach prices to.

Anyway, the hardware guys (and Bob Rogers, philosopher prince of the IBM
brand organization) came up with this solution to a legal problem, not
a technical problem. Bottom line on all of this is that software costs
are the pink elephant in the middle of the room in every conversation
with a customer these days. If they had not done something, whacky as it
was, they would be in even deeper doo-doo than they (and we all) are
now.

CC

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread Don Deese

Tom,

You are correct that these IEAOPTxx parameters (added in z/OS V1R6) 
worked essentially as you described when zAAP was introduced.


However, these descriptions are a bit dated.  The meaning of the 
parameters was changed with z/OS V1R7, and the parameters were 
drastically changed with z/OS V1R8 (and with the ptfs that backfit 
z/OS V1R8 logic to V1R6 and V1R7).


IFACROSSOVER now is ignored completely and is not in IEAOPTxx with V1R8.

Regardless of IFAHONORPRIORITY, zAAP eligible work will execute only 
on zAAP unless Alternate Wait Management (as applied to the zAAP) 
decides that the zAAP needs help to get the work done.  Only if the 
zAAP needs help will z/OS Dispatcher consider zAAP work for 
dispatching to CP.  Even though there is not a ZIPHONORPRIORITY 
parameter, zIIP dispatching works as though YES had been specified.


Additionally, see ZAAPAWMT and ZIIPAWMT in IEAOPTxx of z/OS V1R8 for 
additional possible specifications.


Regards,

Don

**
Don Deese, Computer Management Sciences, Inc.
Voice: (703) 922-7027  Fax: (703) 922-7305
http://www.cpexpert.org
**


At 11:42 PM 4/12/2007, you wrote:

Something else to consider are two IEAOPTxx parameters for WLM control of
zAAP eligible workloads.

IFACROSSOVER controls whether standard CPs will be used for zAAP eligible
work.  If YES (the default), then standard CPs will be used along with zAAP
CPs.  zAAP CPs will be preferred, but standard CPs are not excluded from
consideration based on workload.  If NO, then standard CPs are not
considered for zAAP eligible work unless there are no zAAP CPs present.

IFAHONORPRIORITY controls the workload priority settings for zAAP eligible
work that is executed on standard CPs.  If YES (the default), then all work
on a standard CP will execute based on WLM assigned priority.  If NO, then
all zAAP eligible work that is assigned for execution on a standard CP will
be scheduled at a lower priority than non-zAAP eligible workloads.

I believe these parameters were added with z/OS 1.7.

Tom Moulder


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread John S. Giltner, Jr.
It will work on a z/990 for projecting how much work would be offloaded 
to either zAAPs or zIIPs.


Shane Ginnane wrote:

Sam, is this z9 only ???
What about projecting from z/990 ???

Shane  ...



Nope.  Works fine on prior releases with the zIIP web deliverable and
some APARs for z/OS and DB2 V8.  We have been using this on z/OS 1.7.1
without any zIIP or zAAP installed to monitor eligible time.

-Original Message-
From: IBM Mainframe Discussio

I beleive that PROJECTCPU is only for z/OS 1.8 and up.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-13 Thread John S. Giltner, Jr.
Umm, I was sure that at the NCACMG meeting in March I heard that it was 
for 1.8.  But I guess I mis-heard or got it confused with something else 
that was only available in 1.8.


Knutson, Sam wrote:

Nope.  Works fine on prior releases with the zIIP web deliverable and
some APARs for z/OS and DB2 V8.  We have been using this on z/OS 1.7.1
without any zIIP or zAAP installed to monitor eligible time.

Best Regards, 

Sam Knutson, GEICO 
Performance and Availability Management 
mailto:[EMAIL PROTECTED] 
(office)  301.986.3574 


Think big, act bold, start simple, grow fast...

-Original Message-
From: IBM Mainframe Discussio

I beleive that PROJECTCPU is only for z/OS 1.8 and up.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


zAAP Eligible Work

2007-04-12 Thread Jim Marshall
Went off to one of my ISV vendors and asked if the JAVA coding in their 
product would be run on the zAAP processor I am implementing on my two 
z9BC machines.  The response was 

In order for XXX to use the ZAAP engines, it would have to be started in 
a special enclave eligible for ZAAP processing.  We don't do that.

Does this ring a bell with anyone as being a true statement.

thanks  Jim 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Craddock, Chris
 Went off to one of my ISV vendors and asked if the JAVA coding in
their
 product would be run on the zAAP processor I am implementing on my two
 z9BC machines.  The response was
 
 In order for XXX to use the ZAAP engines, it would have to be
started
 in
 a special enclave eligible for ZAAP processing.  We don't do that.
 
 Does this ring a bell with anyone as being a true statement.

No its completely bogus. That -would- be true for zIIP work, but not for
zAAP. JAVA work runs on a zAAP if there's one present unless you tweak
the JVM to say you don't want it to. 

CC

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Todd Burch
Amen to what Chris said.  

Java is the first implementer of the zAAP engine, and its use is auto-magic.
zAAP is not just for Java apps, it's just that Java was the first app to
get their foot in the door, just as DB2 V8 was first for zIIP exploitation
for certain joins, stored procedures, user defined functions, index related
processing for DB2 utilities, and (phew...) DRDA processing. 

Todd


-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of Craddock, Chris
Sent: Thursday, April 12, 2007 3:18 PM
To: [EMAIL PROTECTED]
Subject: Re: zAAP Eligible Work

 Went off to one of my ISV vendors and asked if the JAVA coding in
their
 product would be run on the zAAP processor I am implementing on my two
 z9BC machines.  The response was
 
 In order for XXX to use the ZAAP engines, it would have to be
started
 in
 a special enclave eligible for ZAAP processing.  We don't do that.
 
 Does this ring a bell with anyone as being a true statement.

No its completely bogus. That -would- be true for zIIP work, but not for
zAAP. JAVA work runs on a zAAP if there's one present unless you tweak
the JVM to say you don't want it to. 

CC

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

-- 
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Edward Jaffe

Todd Burch wrote:

Java is the first implementer of the zAAP engine, and its use is auto-magic.
zAAP is not just for Java apps, it's just that Java was the first app to
get their foot in the door, just as DB2 V8 was first for zIIP exploitation
for certain joins, stored procedures, user defined functions, index related
processing for DB2 utilities, and (phew...) DRDA processing.
  


zAAP runs Java only. And, the interface has not been disclosed.

--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread McKown, John
 -Original Message-
 From: IBM Mainframe Discussion List 
 [mailto:[EMAIL PROTECTED] On Behalf Of Todd Burch
 Sent: Thursday, April 12, 2007 4:39 PM
 To: [EMAIL PROTECTED]
 Subject: Re: zAAP Eligible Work
 
 
 Amen to what Chris said.  
 
 Java is the first implementer of the zAAP engine, and its use 
 is auto-magic.
 zAAP is not just for Java apps, it's just that Java was the 
 first app to
 get their foot in the door, just as DB2 V8 was first for zIIP 
 exploitation
 for certain joins, stored procedures, user defined functions, 
 index related
 processing for DB2 utilities, and (phew...) DRDA processing. 
 
 Todd

Todd,

Are you aware of what other work might be zAAP eligible?

--
John McKown
Senior Systems Programmer
HealthMarkets
Keeping the Promise of Affordable Coverage
Administrative Services Group
Information Technology

The information contained in this e-mail message may be privileged
and/or confidential.  It is for intended addressee(s) only.  If you are
not the intended recipient, you are hereby notified that any disclosure,
reproduction, distribution or other use of this communication is
strictly prohibited and could, in certain circumstances, be a criminal
offense.  If you have received this e-mail in error, please notify the
sender by reply and delete this message without copying or disclosing
it. 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Al Sherkow
I believe it is JAVA applications that are using the IBM JVM (Java Virtual
Machine). If your ISV uses a different JVM then I believe it will not use
the zAAP engine. 

To accumulate data at the address space and service class period level on
how much zAAP and zIIP resources might be consumed if they were configured,
you use a new control specifiable in the SYS1.PARMLIB IEAOPTxx:
PROJECTCPU=YES. This will let you monitor if any applications will use the
specialty engines. 

Al Sherkow, I/S Management Strategies, Ltd.
Consulting Expertise on Capacity Planning, Performance Tuning,
WLC, LPARs, IRD and LCS Software
Seminars on IBM SW Pricing, LPARs, and IRD
Voice: +1 414 332-3062
Web: www.sherkow.com

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Tom Moulder
I have called the zAAP a JPM (JAVA Physical Machine).  How accurate would
that be?  I say that because the processor is no longer a general purpose
processor, so it can not run normal work.

Tom Moulder

-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of Edward Jaffe
Sent: Thursday, April 12, 2007 4:59 PM
To: [EMAIL PROTECTED]
Subject: Re: zAAP Eligible Work

Todd Burch wrote:
 Java is the first implementer of the zAAP engine, and its use is
auto-magic.
 zAAP is not just for Java apps, it's just that Java was the first app to
 get their foot in the door, just as DB2 V8 was first for zIIP exploitation
 for certain joins, stored procedures, user defined functions, index
related
 processing for DB2 utilities, and (phew...) DRDA processing.
   

zAAP runs Java only. And, the interface has not been disclosed.

-- 
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html



-- 
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM


-- 
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread John S. Giltner, Jr.

Al Sherkow wrote:

I believe it is JAVA applications that are using the IBM JVM (Java Virtual
Machine). If your ISV uses a different JVM then I believe it will not use
the zAAP engine. 


To accumulate data at the address space and service class period level on
how much zAAP and zIIP resources might be consumed if they were configured,
you use a new control specifiable in the SYS1.PARMLIB IEAOPTxx:
PROJECTCPU=YES. This will let you monitor if any applications will use the
specialty engines. 


Al Sherkow, I/S Management Strategies, Ltd.
Consulting Expertise on Capacity Planning, Performance Tuning,
WLC, LPARs, IRD and LCS Software
Seminars on IBM SW Pricing, LPARs, and IRD
Voice: +1 414 332-3062
Web: www.sherkow.com



I beleive that PROJECTCPU is only for z/OS 1.8 and up.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Knutson, Sam
Nope.  Works fine on prior releases with the zIIP web deliverable and
some APARs for z/OS and DB2 V8.  We have been using this on z/OS 1.7.1
without any zIIP or zAAP installed to monitor eligible time.

Best Regards, 

Sam Knutson, GEICO 
Performance and Availability Management 
mailto:[EMAIL PROTECTED] 
(office)  301.986.3574 

Think big, act bold, start simple, grow fast...

-Original Message-
From: IBM Mainframe Discussio

I beleive that PROJECTCPU is only for z/OS 1.8 and up.

This email/fax message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution of this
email/fax is prohibited. If you are not the intended recipient, please
destroy all paper and electronic copies of the original message.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Shane Ginnane
Sam, is this z9 only ???
What about projecting from z/990 ???

Shane  ...

 Nope.  Works fine on prior releases with the zIIP web deliverable and
 some APARs for z/OS and DB2 V8.  We have been using this on z/OS 1.7.1
 without any zIIP or zAAP installed to monitor eligible time.
 
 -Original Message-
 From: IBM Mainframe Discussio
 
 I beleive that PROJECTCPU is only for z/OS 1.8 and up.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Tom Moulder
Something else to consider are two IEAOPTxx parameters for WLM control of
zAAP eligible workloads.

IFACROSSOVER controls whether standard CPs will be used for zAAP eligible
work.  If YES (the default), then standard CPs will be used along with zAAP
CPs.  zAAP CPs will be preferred, but standard CPs are not excluded from
consideration based on workload.  If NO, then standard CPs are not
considered for zAAP eligible work unless there are no zAAP CPs present.

IFAHONORPRIORITY controls the workload priority settings for zAAP eligible
work that is executed on standard CPs.  If YES (the default), then all work
on a standard CP will execute based on WLM assigned priority.  If NO, then
all zAAP eligible work that is assigned for execution on a standard CP will
be scheduled at a lower priority than non-zAAP eligible workloads.

I believe these parameters were added with z/OS 1.7.

Tom Moulder


-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of Knutson, Sam
Sent: Thursday, April 12, 2007 10:09 PM
To: [EMAIL PROTECTED]
Subject: Re: zAAP Eligible Work

Nope.  Works fine on prior releases with the zIIP web deliverable and
some APARs for z/OS and DB2 V8.  We have been using this on z/OS 1.7.1
without any zIIP or zAAP installed to monitor eligible time.

Best Regards, 

Sam Knutson, GEICO 
Performance and Availability Management 
mailto:[EMAIL PROTECTED] 
(office)  301.986.3574 

Think big, act bold, start simple, grow fast...

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: zAAP Eligible Work

2007-04-12 Thread Todd Burch
John, publicly, no.  The only workload that I know of today is Java.  Yes,
it's touted as a Java processor, 'cuz that's the only thing using it
today.  Read between the lines from this redbook: 

http://www.redbooks.ibm.com/redbooks/SG246386/wwhelp/wwhimpl/java/html/wwhel
p.htm

Quoting:
When a z/OS logical partition is configured, both CPs and zAAPs are defined
as necessary to support the planned Java and non-Java workloads. zAAPs may
be configured as initially online or reserved for subsequent use by z/OS as
necessary...

If it were only destined to be a java processor, sans any copyright
considerations or politically correct names, I would have expected a name
like zJAP (LOL).  Now, on the other hand, they hit the nail on the head
when they named the IFL.   

It's an application assist processor, or formally, integrated facility
for applications - read between the lines - reduce mainframe TCO - don't
charge for software - more open source leverage...  blah, blah, blah.

Todd







Todd,

Are you aware of what other work might be zAAP eligible?

--
John McKown
Senior Systems Programmer
HealthMarkets
Keeping the Promise of Affordable Coverage
Administrative Services Group
Information Technology

 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.446 / Virus Database: 269.3.0/758 - Release Date: 4/12/2007
11:52 AM
 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html