Radoslaw, Disclaimer: I am a TMM bigot, and I agree 110% (one hundred and ten percent) with the OP.
> > TMM is disk cache backed with real tapes. [Ron Hawkins] I don't quite agree with your definition. TMM can be, and often is migrated and stacked onto to tape but the disk caching does not need to be backed by tape storage. Datasets can live and die in a TMM pool without ever being migrated. I also believe that Management Class provides a greater set of policies for managing dataset residency in the TMM pool than you get with Virtual tape systems. > Advantages: it's free, it fills up (tries to) your tape volumes, quite > easy to set up. [Ron Hawkins] What about: there is always zero mount time in the TMM Pool; there is no limit on number of drives active; Many concurrent readers; Generally faster than TMM appliances; DSORG=PS datasets restored with FCV2, FCV2 manual migration to cheap disk; P-i-T synchronization with DASD for DR > Disadvantages: it may not solve all your needs. [Ron Hawkins] What about: Staging from ML2, No Deduplication, FCV2 manual migration to STORGRUPS on cheap disk; ML2 Tape needs DUPLEX with remote vaults for DR (or similar). > > > My humble opinion: it's good idea to get rid off the tapes in data > processing with the exception for activities like ML2, backups, dumps > and archive. YMMV, but usually we want to have data on DIRECT ACCESS > Storage Device. Direct access is good. > [Ron Hawkins] Agreed Ron ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
