Alan Altmark wrote:
When you bring up z/OS as a guest on the z/VM LPAR, it will report MSU
usage consistent with the share of the CPU it gets from z/VM (SET SHARE).
SET SHARE provides significant granularity in setting processing capacity:
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/hcse4b21/2.494
However, the z/OS 1.9 description for SYSEVENT QVS says:
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/iea2a470/18.14
<IBM Documentation>
QvsVmCapacity is the individual CPU speed multiplied by the number of
online and offline virtual CPUs.
In all cases, the individual CPU speed in based on the MP factor for the
number of online and offline physical CPUs.
</IBM Documentation>
This seems to say that the granularity of processing capacity for z/OS
running as a guest under z/VM is based on the number of apparent CPs
available to the guest. Does z/VM vary the speed of the individual CPs
-- as returned to z/OS by the STSI instruction -- to achieve additional
granularity consistent with SET SHARE?
--
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
