Hi Tony,
I meant that the (DEFINE) SPACE would be similar to a PDS and the
SUBALLOCATED Clusters would be similar to members
in that PDS (SPACE).
Regards,
David
On 2024-05-24 14:26, Tony Harminc wrote:
On Fri, 24 May 2024 at 11:31, David Spiegel <
[email protected]> wrote:
Hi Rex,
VVDSs came with ICF. Before that, VSAM Clusters were ALLOCATED as either
SUBALLOCATION or UNIQUE.
SUBALLOCATION meant that the user ALLOCATED a "cloud" (i.e. DEFINE
SPACE) to hold 1 or more VSAM suballocated Dataset.
Yes, I was about to mention this. All the information for a non-UNIQUE VSAM
dataset was kept in the VSAM catalogue. One advantage in those distant
days of mountable disk drives was that one could allocate or delete a VSAM
dataset with all its details while the disk pack was sitting on the shelf.
This could be seriously useful for production jobs where the packs were
mounted only when the job ran. But of course the down side - as mentioned
here in a slightly different context - was that the disk volume and the
catalogue could get out of sync if the disk failed, and with potentially
much worse results than just some datasets not being catalogued.
It was conceptually similar to a PDS and Members.
I'm not quite seeing that analogy... Have you ever met a PDS with members
on different volumes?
Regards,
David
Tony H.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
