Okay, this is probably the dumbest question of the day, maybe of the month. I am in a discussion with someone and I need some backup for my way of thinking.
If a jobstep reports, for example IEF374I STEP/STEP0010/STOP 2007086.1135 CPU 0MIN 00.02SEC then it is reporting .02 seconds on a single CPU of the box, is that correct? This number would presumably be little changed if the same box went from, say, a 4-way to an 8-way, is that correct? In other words, this is .02 seconds on one CPU, regardless of the number of CPUs in the box, and regardless of the total horsepower (MIPS or MSUs) of the box, right? The only thing that would make this number change significantly would be running it on a box on which each individual CPU was faster or slower -- for example, if each individual CPU was only half as fast it would presumably make it go to about .04 CPU seconds, right? It would not matter if the new box were twice or half as many MSUs, only whether each individual CPU were faster or slower. A "more-MSUs" box would presumably be able to run more such jobs in a day, but each job would still use the same number of CPU seconds, assuming the same speed of CPU? A corollary would be that an 8-way box, for example, has 8 x 86,400 total CPU seconds available to jobs each day (minus overhead and z/OS internal usage, etc.)? Charles Mills ---------------------------------------------------------------------- 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
