Showing papers on "Control reconfiguration published in 1969"

The Distributed Processor Organization

Abstract: Publisher Summary This chapter presents the results of a research study of advanced multiprocessor organizational concepts for future space mission applications, with particular emphasis on a “distributed processor” organization that provides a very high tolerance of failures. It also discusses the requirements and technology investigations. This organization evolved from a study of advanced multiprocessor organizational concepts for future space mission applications. The two leading device oriented technologies that lead the developments in large scale integration (LSI) today are (1) bipolar and (2) metal oxide semiconductor field effect transistors (MOS-FET). The distributed processor organization is diagrammatically presented. The organizational concept exhibits the capability for computational parallelism and achieves high reliability by graceful degradation. A discussion of parallelism within computations to place in proper perspective the parallel computational capability of this organization is presented. The development of this organization and its detailed architecture is also presented. Failure detection and reconfiguration procedures and the design of the cells and the communication buses are also outlined. Some preliminary considerations of the software portions of this organization are discussed.

A teleprocessing approach using standard equipment

Abstract: The techniques described have all been implemented and are working well in an on-line environment assisting control of a vital national facility. The teleprocessing design is not just a technique to combine low-speed and high-speed terminals in one general package; it also provides a basis for development of techniques to provide real-time control facilities on standard general-purpose equipment and support programs. This basis is achieved by the automatic, on-line philosophy of the teleprocessing control program. This program is designed to run 24 hours a day, 7 days a week with minimal operator intervention. No external contact between operators of the central system or the terminal is needed. The foundation is laid for the next development, that of entering direct monitoring signals into the TCP and initiating appropriate calculations or responses. The multiprocessing control by ASP fulfills the reliability consideration of such a system, and it is feasible to build on this to provide automatic reconfiguration after failure, which is a requirement of real-time control. Already, the system described here is capable of reinitializing and restarting itself from an intermittent equipment or program failure, and is running in the environment of a normal batch job processing system.