On Fri, Jun 20, 2014 at 9:01 AM, Elardus Engelbrecht < [email protected]> wrote:
> John McKown wrote: > > >The original spark was from one of our Production Control people. She > basically wanted to look at whether it would be "helpful" in any way to > reduce our active initiator count during certain time frames. > > She must be bored enough to give you unneeded work... > No, we have IT manage constantly "pushing" on us to "reduce cost". If we can reduce MSUs, then we reduce cost. This is just something that she was wondering about. And it sparked a bit of curiosity on my behalf. > > > >If the CEC is overloaded, then it wouldn't hurt to reduce the number of > active initiators. And it might actually help because then individual jobs > would probably finish a bit sooner, there being less multi-job overhead. > > It is true that with reducing the # of jobs (inits) running concurrently, > your CEC load is somewhat lower. Well, at least that worked for us until we > upgraded our z toys. (We had to kick the tooth fairy hard so we can get > toys... ) > All of our toys have been eliminated. We are a "fixed cost" item whereas software is not. So the new mantra is "do as much as you can effectively do BY HAND". Remember the old days of "desk checking" instead of test compiles? It's not quite that bad. > > > >Another thing which is a "biggie" is the 4 hour rolling average MSU. Why? > Because if we are running below our cap, then we are "saving up" MSUs. > > Sort of yes, but the elapsed wall clock time of all those jobs will be > somewhat shorter because of 'less overhead'. > > > >This despite the fact that the z/OS is scheduled to die in Dec 2015. > > That was not the first 'die date'. When was the last 'die date' set last > time? I believe you have posted an earlier date here some ages ago. > Well, one "die" date is later this year. That will likely kill a small part of the z/OS workload. Dec 2015 is what is now published as the "officially planned" die date. But, of course, management will kill z/OS at the first possible instant that they think they can. I.e. when the processing has been reduced to such an extent that what is left can be done "by hand in the user department" (likely going to be some conflict over that). In any case, this is now more a side project of my own. Which I hope will help me learn more R and SQL. And if I can produce a pretty graph, I get the much coveted "attaboy!" award (on genuine recycled letter sized paper) and maybe even a gold star. OOPS, my cynicism is showing again. > > Groete / Greetings > Elardus Engelbrecht > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- There is nothing more pleasant than traveling and meeting new people! Genghis Khan Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
