>In your note below you left out the number of engines as part of the equation >for hardware MSU.
Ah, the MSU is for one processor only, then! Only when I multiply the 74MSU by the 6 processors that we have I arrive at 444 (instead of the quoted 339)., i.e. almost 31% more 'hardware MSU' than 'software MSU'. > It also sounds as if you want to compute the hardware MSU for a series of > models and show then as bars on the same chart as your actual data. The > best way to do that, in my opinion, is to use service units all the way around. > You can do that by converting time back to service units and going forward. This might be an interesting exercise. At this point I am mainly interested in showing the maximum 'capacity' of our basic model. On the other hand, my management knows that we have 339 MSU - if I start using 444, nobody will have a clue what's what. Which might mean that I would have to somehow convert the 444 to relate back to the 339 that are the official MSU rating. As in: We are actually only getting 76% of the full capacity. I could use *that* percentage as 76% of 3600s, meaning the maximum achievable cpu time per 10 minute interval is 2736s. How's that for being convoluted? >Please note that this will not give good numbers or anything close to them if > you change model families, such as from a z10 to a z196. It is only good > within a single family. Agreed. But so far we had been staying in the same family :-) Best regards, Barbara ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
