Re: CFLEVEL 16 and it's structure sizing

2008-11-27 Thread Scott Fagen
On Wed, 26 Nov 2008 22:35:13 -0800, Edward Jaffe
[EMAIL PROTECTED] wrote:

I guess, as long as the rule holds true that higher CF levels nearly
always require the same or more storage, a lazy sysprog can't go wrong
using the values provided by CFSIZER, so long as s/he doesn't run out of
storage. In a short-on-storage situation, closer scrutiny would be
advisable.

Dunno about lazy, but simple economics do play...what's cheaper:  Spending
a few grand on some memory or wasting perfectly good sysprog cycles
rearranging the deck chairs on the good ship Coupling Facility?

Scott Fagen
Enterprise Systems Management

--
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: CFLEVEL 16 and it's structure sizing

2008-11-27 Thread Jan Vanbrabant
Referring to:
Scott Fagen [mailto:[EMAIL PROTECTED]
donderdag , november 27, 2008 05:57 AM
.
From http://www-304.ibm.com/systems/support/z/cfsizer/ :

CFSizer uses your input to calculate the number of structure objects the
structure must contain, based on algorithms provided by the structure owners
that model their use of the structure. Using these results and structure
attribute information provided by the structure owners, CFSIZER THEN SUBMITS
A CALCULATION REQUEST TO A COUPLING FACILITY, USING THE IXLCSP INTERFACE.
The coupling facility itself calculates the structure size required to
accommodate the specified input. Structure sizes therefore correspond to the
CFLEVEL installed on the coupling facility that performs the calculations.
THE CFLEVEL FOR WHICH THE RECOMMENDATIONS WERE CALCULATED IS DISPLAYED ON
THE OUTPUT PAGE AND IS NOT SELECTABLE. Structure sizes are displayed in 
units of 1KB, as they would be entered in your CFRM policy. (emphases added)
Hi Scott,
That means you can't forecast structure size changes with CFSIZER when 
migrating from one level to another when you don't have already a CF at the 
new, higher level.
I came across 'CFSizer Alternate Sizing Techniques'
http://www-304.ibm.com/systems/support/z/cfsizer/altsize.html
Method_2: The SIZER Utility Program (as-is  unsupported!)
In the accompanying document Dave Surman mentions:
However, note that in order for the SIZER utility to be useful for this 
re-sizing purpose, you must already have provided a CF at the new ?uplevel? 
CFLEVEL in your installation, and it must be configured and active in your 
sysplex, with connectivity to the system on which the SIZER utility is being 
run.
Conclusion: it seems there's no way to estimate future structure sizing 
requirements when having no access to a CF of the new/higher CFlevel?!
Seems a bit strange to me.
Is there no ROT when going to CFlvl 16 (as there once used to be one when 
migrating to CFlvl 12 if I recall well)?
Jan


--
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: CFLEVEL 16 and it's structure sizing

2008-11-27 Thread Edward Jaffe

Scott Fagen wrote:

On Wed, 26 Nov 2008 22:35:13 -0800, Edward Jaffe
[EMAIL PROTECTED] wrote:

  

I guess, as long as the rule holds true that higher CF levels nearly
always require the same or more storage, a lazy sysprog can't go wrong
using the values provided by CFSIZER, so long as s/he doesn't run out of
storage. In a short-on-storage situation, closer scrutiny would be
advisable.



Dunno about lazy, but simple economics do play...what's cheaper:  Spending
a few grand on some memory or wasting perfectly good sysprog cycles
rearranging the deck chairs on the good ship Coupling Facility?
  


I agree. And, for the record, I consider myself to be a lazy sysprog. 
I have better things to do than micro-manage system configurations. I 
over-allocate everything. To that end, our default region size is 256MB 
and default MEMLIMIT is 2G!


--
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: CFLEVEL 16 and it's structure sizing

2008-11-27 Thread Norman Hollander on DesertWiz
And please remember to have enough memory for failover from other 
CFs.  Same for enough processor power.  Most folks believe a CF 
won't fail.  If you are brave enough to believe that for your 
mission critical data sharing applications, then no problem.  BUT...

-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of Scott Fagen
Sent: Thursday, November 27, 2008 SYSN 07:58 AM
To: IBM-MAIN@bama.ua.edu
Subject: Re: CFLEVEL 16 and it's structure sizing

On Wed, 26 Nov 2008 22:35:13 -0800, Edward Jaffe
[EMAIL PROTECTED] wrote:

I guess, as long as the rule holds true that higher CF levels nearly
always require the same or more storage, a lazy sysprog can't go wrong
using the values provided by CFSIZER, so long as s/he doesn't run out of
storage. In a short-on-storage situation, closer scrutiny would be
advisable.

Dunno about lazy, but simple economics do play...what's cheaper:  Spending
a few grand on some memory or wasting perfectly good sysprog cycles
rearranging the deck chairs on the good ship Coupling Facility?

Scott Fagen
Enterprise Systems Management

--
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

--
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


CFLEVEL 16 and it's structure sizing

2008-11-26 Thread Jan Vanbrabant
Hi,

https://www-304.ibm.com/jct03001c/systems/z/advantages/pso/cftable.html
just mentions CFLEVEL 16 is (or will be ?) available 
on 2098 z10 BC EC N10964 (Driver 76G) MCL 001
It doesn't even contain some info on what CFlvl16 is about?

Is there already CFlvl16 information available?

What about structure sizing?
No techdoc yet? No ROT yet? 
Will the CFSIZER be the only source of info about sizings?
Has the CFSIZER already been updated? 

Jan




--
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: CFLEVEL 16 and it's structure sizing

2008-11-26 Thread Vernooy, C.P. - SPLXM


Jan Vanbrabant [EMAIL PROTECTED] wrote in message
news:[EMAIL PROTECTED]...
 Hi,
 

https://www-304.ibm.com/jct03001c/systems/z/advantages/pso/cftable.html
 just mentions CFLEVEL 16 is (or will be ?) available 
 on 2098 z10 BC EC N10964 (Driver 76G) MCL 001
 It doesn't even contain some info on what CFlvl16 is about?
 
 Is there already CFlvl16 information available?
 
 What about structure sizing?
 No techdoc yet? No ROT yet? 
 Will the CFSIZER be the only source of info about sizings?
 Has the CFSIZER already been updated? 
 
 Jan
 
 
 

I received information from the ITSO roadshow last week, but I don't
know where the official announcement is.

Some hilites:
- Performance enhancementes for System Managed Duplexed Structures
- Shared Q enhancements: eliminate false scheduling
- Structure size increases, not for CFlvl 16 itself, but as a
positioning for CFlvl 17.
- CFSIZER is now running at CFlvl 16 level.

Kees.
**
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 [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


Re: CFLEVEL 16 and it's structure sizing

2008-11-26 Thread Edward Jaffe

Jan Vanbrabant wrote:

Has the CFSIZER already been updated?
  


Yes. The CFSIZER has been updated for CF level 16.

How do people running earlier CF levels do their sizing? I did not see 
an entry field for the CF level...


--
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: CFLEVEL 16 and it's structure sizing

2008-11-26 Thread Scott Fagen
On Wed, 26 Nov 2008 07:58:34 -0800, Edward Jaffe
[EMAIL PROTECTED] wrote:
-snip-
How do people running earlier CF levels do their sizing? I did not see
an entry field for the CF level...

That's an exercise left to the reader:

From http://www-304.ibm.com/systems/support/z/cfsizer/ :

CFSizer uses your input to calculate the number of structure objects the
structure must contain, based on algorithms provided by the structure owners
that model their use of the structure. Using these results and structure
attribute information provided by the structure owners, CFSIZER THEN SUBMITS
A CALCULATION REQUEST TO A COUPLING FACILITY, USING THE IXLCSP INTERFACE.
The coupling facility itself calculates the structure size required to
accommodate the specified input. Structure sizes therefore correspond to the
CFLEVEL installed on the coupling facility that performs the calculations.
THE CFLEVEL FOR WHICH THE  RECOMMENDATIONS WERE CALCULATED IS DISPLAYED ON
THE OUTPUT PAGE AND IS NOT SELECTABLE. Structure sizes are displayed in
units of 1KB, as they would be entered in your CFRM policy.  (emphases added)

I'd suggest honing your IXLCSP skills...start with (mind any line breaks
that may be inserted):

http://publib.boulder.ibm.com/infocenter/zos/v1r9/index.jsp?topic=/com.ibm.zos.r9.ieai600/usecsp.htm

Scott Fagen
Enterprise Systems Management

--
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: CFLEVEL 16 and it's structure sizing

2008-11-26 Thread Edward Jaffe

Scott Fagen wrote:

On Wed, 26 Nov 2008 07:58:34 -0800, Edward Jaffe
[EMAIL PROTECTED] wrote:
-snip-
  

How do people running earlier CF levels do their sizing? I did not see
an entry field for the CF level...



That's an exercise left to the reader:

From http://www-304.ibm.com/systems/support/z/cfsizer/ :

CFSizer uses your input to calculate the number of structure objects the
structure must contain, based on algorithms provided by the structure owners
that model their use of the structure. Using these results and structure
attribute information provided by the structure owners, CFSIZER THEN SUBMITS
A CALCULATION REQUEST TO A COUPLING FACILITY, USING THE IXLCSP INTERFACE.
The coupling facility itself calculates the structure size required to
accommodate the specified input. Structure sizes therefore correspond to the
CFLEVEL installed on the coupling facility that performs the calculations.
  


I guess, as long as the rule holds true that higher CF levels nearly 
always require the same or more storage, a lazy sysprog can't go wrong 
using the values provided by CFSIZER, so long as s/he doesn't run out of 
storage. In a short-on-storage situation, closer scrutiny would be 
advisable.


--
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