On Thu, 24 Mar 2022 19:49:24 +0000, Michael Watkins wrote:

>    ... A quick search of the IBM documentation doesn't seem to provide a 
> conclusive answer. Anyone got a clue? 
>
I'm confident you'll get a knowledgeable answer.

I wonder, which is better:
o More single-volume data sets for greater I/O overlap?
o Fewer multi-volume data sets using less buffer space?

SORT knows best; let it choose.

Or using only 64-bit virtual for workspace?  The technique
should be designed with LoR to optimize paging.

Do virtual disks do anticipatory reads?

Is bus bandwidth saturation a concern?

-- 
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to