Thus far we have 1.13 installed on systems at our DR site. The bulk of the daily work that goes on on these systems, is DFDSS restore from full volume disk backup datsets that are FTP'ed to that site and we have observed these restore jobs running in about 50% less wall clock time than before the upgrade. Could the change from EXCP to BSAM account for this also?
Dana On Thu, 16 Aug 2012 17:42:29 -0500, Steve Thompson <sthomp...@us.ibm.com> wrote: >__________________ > >We have seen massive savings - primarily in wall-clock savings - but from >what I understand (2nd and 3rd hand): > >3490s have a max blksize of 64k; 3590s have a max blksize of 256K >If you use virtual tapes - make sure the virtual tapes are defined as >3590s to get the best LBI benefit > ><SNIPPAGE> > >Please forgive me but I am having issues with my email system, so headers >got lost. > >At any rate, at 1.12 the DEVTYPE (I think that was the macro) gets updated >to allow 3490s to go LBI at 256K blocks. > >If you are running Connect:Direct for z/OS, you will need the equivalent >PTF. > >Certain ISV software may not handle the change from 65K max blksize for >the 3490s. > >Be careful if you are going to have one system at 1.12 (or greater) and >another system below 1.12, because you will run into problems with the >lower system reading the tapes. > >There are exits for the 1.12 level in handling this. Look at the ADRDSSU >manuals with regards to those exits. > >Regards, >Steve Thompson >Connect:Direct for z/OS >IBM -- Irving TX Lab > >---------------------------------------------------------------------- >For IBM-MAIN subscribe / signoff / archive access instructions, >send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
