[email protected] (Paul Gilmartin) writes: > If gaps and other unused space are virtualized, there is no reason for SDB > ever to choose a BLKSIZE other than 32760 (or nearest multiple of LRECL). > I suspect there are various implementations of RAID: some may virtualize > unused space; others keep images of entire tracks, however sparsely used. > > I tried the experiment I suggested: I created 10 mebers of 350 records. > With RECFM=FB,LRECL=80,BLKSIZE=27920 (chosen by SDB) the data > set used 10 tracks. When I forced to BLKSIZE=24000, it used 7 tracks. > > Admittedly a deliberate worst case.
it isn't just RAID ... aka in ancient past there could even have been RAID on real CKD disk ... however, there hasn't been any *real* CKD disks made for decades ... all CKD being simulated on industry standard disks that are effectively fixed-block architecture. big transition now is the transition from fba-512 to fba-4096. http://en.wikipedia.org/wiki/Advanced_Format RAID just adds another layer on top of FBA ... further increasing the distance of the artificial CKD simulation from anything remotely considered real hardware. http://en.wikipedia.org/wiki/RAID -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
