On Tue, 11 Aug 2009 08:04:13 -0500, Staller, Allan <[email protected]> wrote:
>>Question 1 (Is this correct?) >>If you specify MIGRATION(REUSE) an ML2 tape is not selected for >>duplexing. If you want an ML2 to be duplexed, you need to either >>specify SETSYS PARTIALTAPE(MIGRATION(MARKFULL)) or generate a list >>of partialtapes >>LIST TTOC SELECT(ML2 NOTFULL) ODS(OUTPUT.DATASET) and mark >>them full so that TAPECOPY will automatically duplex them. > >Answered in another post. > >>Question 2 (Is this also correct?) >>If the above is correct, then by marking an ML2 volume full before it >is >>actually full will probably triple the number of scratch tapes >required. >>An extra tape for each duplex copy and since all of a tape will not be >>used, more tapes will be needed. > >Recycle is your friend here! > >This is the way it really works: >Data is written concurrently to both the primary and duplex copy of ML2. >When either the primary or duplex becomes full, (1st to EOV wins) both >are so marked and another duplex pair is created. The partialtape >parameter enables the recycle to be performed immediately. Of course you >may not want to do this since it will foul up Fast Subsequent Migration. > >HTH, > Thanks for the info, every tidbit helps. We currently defer the recyclying of tapes to the weekends when processing is slower. Thanks again, Gil. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
