Suppose you were to prepare a small program to write fixed-length
records that were filled with binary zeros. Make the LRECL slightly
larger than half-track size so that only one record will fit per track.
Art, you got me thinking. As you pointed out, the bad side of your idea
is that it does use some backend storage for each track of zeros. My
reference for the STK SVA (successor to the RVA) says that it uses
backend in 2k increments, not 16K, but 16K might be right for the RVA.
Anyways, earlier in this thread someone suggested doing an ICKDSF INIT
on the volume and I responded that was a bad idea because it would use
backend storage. But I think I was wrong
I believe that if a track in the RVA contains only a home address and a
standard R0 (with a CCHH matching the real CCHH and a data length of
zero), no data records on the track at all, then this track is NOT
stored in the backend; it is just a table entry indicating it is an
empty track.
A ICKDSF Medial INIT (INIT with NOCHECK and VALIDATE specified) will do
exactly this to every track on the volume, which will release the
backend storage previously associated with any tracks on the volume.
Of course, this can be done only to an offline volume from which all
data has been moved elsewhere.
--
Bruce A. Black
Senior Software Developer for FDR
Innovation Data Processing 973-890-7300
personal: [EMAIL PROTECTED]
sales info: [EMAIL PROTECTED]
tech support: [EMAIL PROTECTED]
web: www.innovationdp.fdr.com
----------------------------------------------------------------------
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
