On Wed, 5 Dec 2007 16:36:30 -0800, Edward Jaffe wrote:
>Paul Gilmartin wrote:
>> Ummm. I think the platter size is pretty much maxed out to the
>> capacity of CC. But the unused capacity of HH should allow for
>> thousands more platters than 15.
>>
>
>And that's four bits, which is a lot! Disk addresses are currently of
>the form 'cc0hr' where '0' is an unused nibble for 3390 geometry. If the
>current max disk size is 54GB, then 16 times that much would be ...
>
ITYM "bytes", not "nibbles". I see:
6.7 "z/OS V1R7.0 DFSMSdfp Diagnosis"
DESCRIPTION OF THE F7 FORWARD POINTER
135 (87) CHARACTER 5
DS7PTRDS POINTER (CCHHR) TO NEXT FORMAT 7 DSCB OR ZERO
135 (87)
* DS7CCPTR TWO BYTES OF CYLINDER DATA (CC)
* DS7HHPTR TWO BYTES OF HEAD DATA (HH)
* DS7RPTR ONE BYTE OF RECORD DATA (R)
* DS7END END OF THE DSCB 7 MAPPING
So, not 54 GB times 16, but 54 GB times 4096.
Have I missed something?
"No one will ever need more than 640 KiB."
"No one will ever need more than 32 MiB."
"No one will ever need more than 2 GiB."
etc.
-- 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
