> Also, our current limitations notwithstanding, a snippet of a TASID > ====================================================== > Jobname | Procname Stepname CPU Time Storage > ------------------------------------------------------ > UTSO001 ... TPROC001 ... EXEC ... 21:53.34 90.2M > ====================================================== > > How is it that this user has apparently been able to allocate and use > 90.2M storage, considering:
I don't know exactly what TASID is telling you or where it gets that data, but in a completely vanilla system it is pretty easy to see how it could happen. Your REGION request was 4M. That means you will get 4M below and given that you say you don't have any IEFUSI second-guessing the programmer, you would get the system default region above. IIRC that is 32M, but I haven't looked in a long while. So you could get at least 36-odd meg just from user data and application code. That might or might not be concurrently allocated and it might or might not show up as working set, depending on how storage-rich your processor is. If you've got memory to burn, the system isn't going to reclaim any pages unless/until it needs them. So where would the "extra" memory come from? Authorized subpools don't count in the region, so if you're using some vendor tool like a performance monitor you could easily be using a large amount of storage above the region limit and if you're fat on storage that could also show up as working set. You may also have some number of garden variety dataspaces in use. None of those are controlled (at least not in your environment given no USI) So its not hard to believe 90M at all. CC ---------------------------------------------------------------------- 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
