>I calculated my minimums and caps based on a percentage of the MSU's available >on the machine. The last time we upgraded to the z9, I had to adjust all the minimums and maximums based on the new machine. So going to a percentage scale will fit nicely in that scheme. And I never have to change them again based on a machine upgrade.
I guess that's were I disagree. I use resource classes based on a min/max of service, rather than processor capacity. I set it up, now and never change it. Regardless of the processor capacity, my TEST workload will never get anymore than it had. With my choice the test workload will never get any more resource. With yours, it will grow with each upgrade. Plus, it doesn't work with multiple CEC's (differing capacities) in the same SYSPLEX. Plus, I have a question. Why do you want to give more resource to the unimportant workloads, just because you've done an upgrade? - Too busy driving to stop for gas! ---------------------------------------------------------------------- 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
