Re: JCL, The Design of Design

John Levine wrote:

According to Scott Lurndal <slp53@pacbell.net>:

The worst parts of JCL were "DD" cards, manual track allocation,
and the lack of any form of real filesystem.   PDS do not count.

 
OS had named files (which they called datasets) and a catalog that
said which disk or tape it was on, so you could refer to SYS1.FOOLIB
and it would find it for you.

Agreed.  But by "real file system", Scott meant a hierarchical system
such as we are used to today with Windows and Unix, then neither OS nor
Exec 8 had that.

I agree the explicit track and cylinder allocation was painful. IBM
outsmarted themselves with CKD disks that put the record boundaries in
hardware

Technically block boundaries.  Anyone who did unblocked 80 byte card
image data on a disk diserved the poor space ysage and performance he
deserved.

and did the key search for indexed files in the disk
controller.

It was worse than that, even for non keyed datasets, as each key or
count field for a block came under the read head, the controller sent
the indication of a match or not back to the host channel, which, if it
wasn't a match, reissued the search command.

That was fine on a 360/30 which was too slow to do
anything else (the CPU and channel shared the microengine and the CPU
basically stopped during an index search) but very inefficient on
larger machines.
 
Later on VSAM handled disks with fixed block sizes and used B-trees to
do index searches but by then the cruft was all there.

Yes, but even early VSAM incurred the overhead until the Extended CKD
stuff (i.e. define extent and locate commands) came into common use.




--
 - Stephen Fuld
(e-mail address disguised to prevent spam)