>One curious side effect is that we seem to be getting higher SIO rates on some >jobs / stcs in SDSF.
You have to be careful with real-time monitoring and looking at rates. It's the number of I/O's since you last hit enter divided by the number of seconds since you last hit enter. A more consistent measure would be either an RMF Interval, or auto updates under SDSF, which can only be done at a TSO Command Prompt (iirc). Or OMEGAMON with AUPON. >Is this just because we can now do more SSCH instructions in a given time frame? It's possible, but you need (as I said) a longer monitoring interval than just hitting enter. In theory, especially if you had CPU-Bound jobs that there will be more SIOs in a given interval, but there could be less, if previously CPU-Bound jobs are dependent on others of the same type, and they are starting earlier. You really can't get an effective measure on a job by job basis (imo). You should look at RMF and the SIO's per CPU busy. But, even that can change due to other factors, such as adding more memory (which I usually do at CPU uipgrade time). - 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
