How about just pick big (ish) number and keep an eye out for mount waits/failed jobs (i'm assuming you'll get a mount wait if the number isn't big enough - that may not be correct)
Not scientific I know - but much easier :) Steven On 5 December 2012 19:09, Arbogast, Warren K <[email protected]> wrote: > Richard, > Thank you for that detailed answer. It sounds difficult to track and map, > and possibly difficult to correlate increases and decreses of > numopenvolumesallowed with system throughput. TSM likes no easy answers. > > Best wishes, > Keith >
