by Conrad Weisert, April 2, 2017
©Information Disciplines, Inc.
NOTE: This document may be circulated or quoted from freely,
as long as the copyright credit is included.
Just three years ago we paid a nostalgic tribute to the IBM 2314, a direct-access storage device introduced in 1965. By comparison with its predecessors it was amazingly small taking up a reasonable quantity of floorspace and supporting removable media that a human operator could easily carry. In that article we also noted the current growing availability of much smaller high-capacity devices.
Progress in storage technology continues. A recent ComputerWorld article (part 1 of 2) surveys the amazing growth in capacity, reduction in physical size, and lowering of per-byte cost in storage technology.
The storage subsystems of 1960s and 1970s large-scale operating systems supported global cataloguing.1 When a program created a new data-set2 the operating system would make an entry in its catalog, associating the data-set name with its physical location. Subsequent retrieval of the data-set was by name only, specified through the Job-Control Language. Neither the jobs themselves nor the users who submitted them needed to know where the desired data set was stored.
The data-set location, recorded in the catalog, could be either on a permanently mounted device or on an off-line volume, which the operator would be instructed to fetch and mount. That scheme worked not only for disks and other direct-access devices but also for old-fashioned tape reels.3
The catalog, of course, had to be permanently mounted so that the operating system could access it quickly. Unfortunately, maintaining a global catalog of every data-set in an organization became unwieldy as:
Last modified April 2, 2017
Return to IDI home page