FAULT-TOLERANT COMPUTING:A PREAMBLE FOR ASSURING VIABILITY OF LARGE COMPUTER SYSTEMS

摘要

This paper describes why large computer systems need to be more fault-forgiving, or fault-tolerant, in order to be viable. The need for fault-tolerant computing is addressed from the viewpoints of (1) why it is needed, (2) how to apply it in the current state of technology, and (3) what it means in the context of the Phoenix computer system and other related systems. To this end, the value of concurrent error detection and correction is described from the viewpoints of (1) user protection, (2) program retry, and (3) repair. The technology of algebraic codes to protect memory systems and arithmetic codes to protect arithmetic operations is discussed.