On Tue, 23 Mar 2021 16:54:57 +0100, Radoslaw Skorupka wrote:
>
>>> However when you specify records or kilobytes, the system assumes the
>>> following:
>>> a) you have only inkling about mainframe DASD
>>>
>> Wrong!  It forces me to know more than I care to.
>
>Agreed. However that was the intention. Personally I discourage students
>from using AVGREC. It is counterintuitive IMHO, especially when you have
>to record sizes - in LRECL and SPACE.
>Of course TRK/CYL require some calculation to know how many records... -
>it is the same like in KB or MB.
>
That hypothetical naive programmer shouldn't know TRK/CYL.  They're
no longer real anyway.

>>> b) you want such and such number or kilobytes, megabytes or records. AT
>>> LEAST.
>>>
Suppose I specify RECFM=VBS,LRECL=X,BLKSIZE=0.  What should I say
in SPACE, etc. to get 1GB.  Without relying on device specifics.

>> A PITA when I'm generating JCL with a program.  Why does IBM choose
>> to afflict me with 20th Century, 2311 mentality?
>
>Agreed. I would vote for SPACE=(GB,(10,5))  and TB, and MB, and PB of
>course.
> From the other hand I'm so used to TRK and CYL so I don't feel any pain
>with that. Especially I saw the last non-3390 DASD in... 
>
Don't think of the 3390 geometry as divinely ordained.  A recurrent legend
here is that IBM once considered FBA, but deemed the software support
cost prohibitive.  So real FBA emulates CKD.

Some decades ago I attended a presentation where IBM touted an online
service.  The default subscription included 1 MB (sic) DASD storage.
Question from the audience: "How many cylinders is that?"  I forget
the answer.  It should have been, "It depends ..."

-- gil

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

Reply via email to