> -----Original Message----- > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > Behalf Of Peter Hunkeler > Sent: Wednesday, April 01, 2015 12:52 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Why is the boundary between TRK- and CYL managed space at cylinder > 65519/20? > > I wonder why the boundary between track managed and cylinder mamanged space is > between cylinders 65519 snd 65520. And why is the space unit 21 cylinders when it comes > to the cylinder mamange space? > I understand there there are two things that are different between those two areas (see also > the excerpt from a DFSMS manual below): > a) For track managed space, allocation information is kept as x'cccchhhh', so a half word for > the cylinder and track (head) numbers, each. For cylinder managed space it is x'ccccCCCh', > i.e. 28bits for the cylinder and only 8 bits for the head address. > b) For track managed space, allocation is in units of one track, for cylinder managed it is in > units of 21 cylinders. > > > 65535 / 21 is 3120.73120 * 21 is 655203121 * 21 is 65541 > > I guess from the above calculations that 65520 was chose so that no space is "lost" between > the maximum tracked mananged address (theoretically 65535) and the first cylinder > managed address 65541. This assumes that there are no non-EAV volumes out there which > have 65535 cylinder (I think this yould be possible under VM, is it?). > > I don't have an idea why "21 cylinders" was chosen as the cylinder managed unit. Anyone?
Could it have anything to do with how the FBA hard disk emulates a CKD disk volume? Does 21 cylinders equate to a nice round, or otherwise convenient, number of sectors? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
