>>the slower cp is the measure and not the faster cp. >I would have said that "the standard CP is the measure and not the >specialty engine" but of course in today's world the two statements end up >the same since the standard CP is the slower one when there is a >difference.. I know that I tend to formulate things a little - shall we say - sloppily?
>Perhaps this was taken out of context (it might even have been written >before there were zAAPs and zIIPs)t, but to get the SU capacity of a CEC, >you need to know also how many there are of each type of CPU and factor in >the speed differential. I understood what Allan meant. I have looked at the 'srm constants' that govern the SU conversion factor and hence the rating. (or is that the other way around, too?) In any case, we look at all types of CPs separately - IFLs for VM, ZIIP separate from GCPs, and other than me occasionally, nobody looks at how busy the CF processors are. I agree that it is hard to figure out the z/OS part of the capacity, precisely because the ZIIP is faster than the general cp. That this would translate into more cpu seconds per wall clock second explains why the ziip times never really matched with runtimes of the jobs - we've noted discrepancies before. But I wonder: Does VM have anything like a SU rating? Or in other words, do SUs figure into the IFL capacaity? And on the same box, given that IFLs are faster, does that mean that the VM monitor (which reports in seconds) give me more cpu seconds per wall clock second, too? I report percentages for VM and its guests, so I wouldn't have noted anything, but for our accounting, I am using the total cpu seconds (since I couldn't find anything else way back when I first established this reporting). Best regards, Barbara ---------------------------------------------------------------------- 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
