A scheduling philosophy for multiprocessing systems
Communications of the ACM
| Hi-index | 0.00 |
A third generation computer seems to cause as many problems as it solves; not because it is difficult to program or too problem directed---quite the contrary. The problems arise because such a computer lends itself so willingly to all applications---realtime data acquisition, process control, scientific calculations, bookkeeping and conversational time-sharing. In a nuclear physics laboratory, there are enough imaginative people interested in each of these subjects that eventually all are implemented with some success. The central problem, then, is to develop an operating environment compatible with open-ended development of any or all types of computer usage. Ideally, one seeks a standard operating system providing the framework and resources to aid all such development.