On Mon, 17 Oct 2005 14:44:41 -0500, Dave Danner <[EMAIL PROTECTED]> wrote: >On Mon, 17 Oct 2005 13:41:00 -0400, Knutson, Sam <[EMAIL PROTECTED]> wrote: -snip- >>This provided over an 80% reduction from the previous levels of CONSOLE >>CPU use during our peak WLC charging hours! -snip- >Mr. Fagen can >correct me if I'm wrong, but I thought .FORNSSI was pretty much just a band- >aid solution until the more robust redesign in OA09229 was available. If >you still see an appreciable difference between turning .FORNSSI on and off >after OA09229 I'd certainly like to know about that.
It's true that OA09229 will fix the lion's share of the additional overhead caused by sending messages around the sysplex (it eliminates the automatic multicast of all messages), but if you do use the ability to route messages around the sysplex, OA08482+.FORNSSI *NONE will reduce CPU in the CONSOLE address space by preventing these messages from going to subsystems that will not have any interest in processing them. I hesitate to make this a 'best practice' since I am not aware of any study to determine if it always safe to turn this switch on. I'm sure there are subsystems that want to see foreign messages (i.e. focal point automation). Scott Fagen z/OS Core Technology Design IBM Poughkeepsie ---------------------------------------------------------------------- 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
