Setting Coupling Facility VOLATILE: NO

[Knutson, Sam]

Hi,



I have been looking at Logger performance as we have increases
significantly the amount of data going into Logger in the last few
years.

One thing I am considering in conjunction with our next CEC installation
is to have the z/OS team set the ICF partition in the new CEC as
VOLATILE: NO.

With a non-volatile CF in another failure domain (CEC) it looks like I
could improve throughput and reduce I/O on some of my heavily used
logstreams.

We have redundant power connection and triple redundant power
architecture in the data center.

It seems that if you have a reliable UPS infrastructure and redundant
connections from the z box to different PDU's this is reasonable.



Does anyone order the internal battery on a CEC? Does anyone set a
standalone CF or ICF partition as VOLATILE: NO without an internal
battery?

Thanks,

Sam Knutson, GEICO
Performance and Availability Management
mailto:***@geico.com
(office) 301.986.3574

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













====================
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 ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Lizette Koehler]

I seem to recall that DB2 has issues with VOLATILE but do not remember what
they are. Will you have DB2 running on them?

Lizette



[>] Snip

I have been looking at Logger performance as we have increases
significantly the amount of data going into Logger in the last few
years.

One thing I am considering in conjunction with our next CEC installation
is to have the z/OS team set the ICF partition in the new CEC as
VOLATILE: NO.

With a non-volatile CF in another failure domain (CEC) it looks like I
could improve throughput and reduce I/O on some of my heavily used
logstreams.

We have redundant power connection and triple redundant power
architecture in the data center.

It seems that if you have a reliable UPS infrastructure and redundant
connections from the z box to different PDU's this is reasonable.



Does anyone order the internal battery on a CEC? Does anyone set a
standalone CF or ICF partition as VOLATILE: NO without an internal
battery?
[>] UnSnip

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

[Vernooy, C.P. - SPLXM]

"Knutson, Sam" <***@GEICO.COM> wrote in message news:<***@GP2K0084V3.GEICO.corp.net>.
..

Post by Knutson, Sam
Hi,
I have been looking at Logger performance as we have increases
significantly the amount of data going into Logger in the last few
years.
One thing I am considering in conjunction with our next CEC

installation

Post by Knutson, Sam
is to have the z/OS team set the ICF partition in the new CEC as
VOLATILE: NO.
With a non-volatile CF in another failure domain (CEC) it looks like I
could improve throughput and reduce I/O on some of my heavily used
logstreams.

Where and why do you forsee the improvements?
AFAIK a CF in the same CEC, connected with internal microcode emulated
CF links, is still much faster than external links to a CF in another
CEC. So you mean you can improve performance by moving to an external
CF?

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 ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Skip Robinson]

I'm also curious about performance differences vis a vis volatility. In the
early days of parallel sysplex, some exploiters used the volatility
indicator when deciding whether a CF was acceptable for structure
allocation. I haven't heard much of that in years. We recently upgraded a
couple of CECs and therefore had to set up new ICF LPARs. Even my IBM rep
recommended setting them nonvolatile because we have several backup power
sources.

We stopped ordering the internal battery long before we moved from
standalone to internal CF.

.
.
JO.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
626-302-7535 Office
323-715-0595 Mobile
***@sce.com



"Vernooy, C.P. -
SPLXM"
<***@KLM To
.COM> IBM-***@BAMA.UA.EDU
Sent by: IBM cc
Mainframe
Discussion List Subject
<IBM-***@BAMA.UA Re: Setting Coupling Facility
.EDU> VOLATILE: NO


07/03/2008 06:17
AM


Please respond to
IBM Mainframe
Discussion List
<IBM-***@BAMA.UA
.EDU>








"Knutson, Sam" <***@GEICO.COM> wrote in message news:<***@GP2K0084V3.GEICO.corp.net>.
..

Post by Knutson, Sam
Hi,
I have been looking at Logger performance as we have increases
significantly the amount of data going into Logger in the last few
years.
One thing I am considering in conjunction with our next CEC

installation

Post by Knutson, Sam
is to have the z/OS team set the ICF partition in the new CEC as
VOLATILE: NO.
With a non-volatile CF in another failure domain (CEC) it looks like I
could improve throughput and reduce I/O on some of my heavily used
logstreams.

Where and why do you forsee the improvements?
AFAIK a CF in the same CEC, connected with internal microcode emulated
CF links, is still much faster than external links to a CF in another
CEC. So you mean you can improve performance by moving to an external
CF?

Kees.

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

[Barbara Nitz]

Oh yes, the good old volatility chapter:

Logger determines single point of failure for a logger connection to a structure and attempts to allocate (if possible) so that the connection is non-volatile. Single point of failure in turn determines if a log stream is duplexed (using duplexmode(cond), which is the default).

Sam,
1. what duplexmode are you using?
2. when you specify non-volatility for the CF, will your log stream be in a single point of failure situation? (Remember that CF volatility is only one criteria)

If the answers are 1 COND and 2 NO, then you may see less cpu consumption because duplexing is not done anymore. Otherwise, I don't see how it will help performance. I think.

I have no clue if we have an internal battery :-)

Best regards, Barbara