You shouldn't make the maximum size too large.

There is a somewhat new health check called XCF_CF_STR_POLICYSIZE that "Check 
that structures in the CFRM active policy do not have too large a difference 
between the value specified for INITSIZE and |the value specified for SIZE. All 
specifications of INITSIZE in the |active or pending CFRM policy should 
indicate an initial structure size of at least half the maximum structure size 
(as determined by the SIZE specification)."

Alan Schwartz
ITO Global Services Operations and Engineering
Xerox Business Services, LLC
1500 Towerview Rd.
Eagan, MN. 55121-1346

p.  612.266.3150
m. 651.274.5819
f.   612.266.3196

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Bill Neiman
Sent: Friday, April 12, 2013 11:32 PM
To: [email protected]
Subject: Re: CFRM policy size

You should use the size that corresponds to your actual workload and 
configuration.  If you use a structure that's too small, you'll get fewer lock 
table entries and may encounter more false contention (when distinct resources 
hash to the same lock table entry) and degraded performance.  If you use a 
structure that's too large, you'll waste some CF storage.  In the case of GRS, 
nothing disastrous is likely to happen, and you can always adjust the size on 
the fly by updating the CFRM policy and rebuilding the ISGLOCK structure.  
CFSizer provides a starting point.  It's up to you to determine what size works 
for your installation.

Bill Neiman
Parallel Sysplex development
IBM Poughkeepsie

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
[email protected] with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to