> All page D/S should be the same size (along with any to add later). > The ASM uses the smallest times the number attempting to (in > general) fill to a max of 65% (iirc).
That's just crazy talk. While there might be some load balancing related motivations for using similar sized page data sets, ASM most certainly uses the full capacity of all of the data sets when calculating auxiliary storage shortage thresholds. > > Yeah, we've been told to do the same thing, but I need to know why > > this suddenly happened. All we did was move our page data sets to the > DS8800. > > Did the move preserve the number of page data sets, and the size of > the page data sets? Was the total amount of page data set space the > same before and after the move? > No, at least one of the page data sets was larger during the initial > migration. Also, as the page datasets fill up they'll add additional > ones of varying sizes. The number of available page data set slots is one (of many) things considered by the STGTEST SYSEVENT, which I would guess is used by DFSORT. Increasing the number of available slots might change the results returned by STGTEST in a way which might allow DFSORT to decide to be more aggressive about using storage. Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
