Fault-Tolerant Computing: A Introduction

THE field of fault-tolerant computing is concerned with the analysis, design, verification, and diagnosis of computing systems that are subject to faults. A "computing system," in this general context, can be a hardware system, a software system, or a computer which includes both hardware and internal software. A "fault" can reside in either hardware or software and can occur in the process of designing and implementing the system, or in the process of using the system once it is implemented. Major areas of technical interest include: 1) the design and analysis of computers which are able to execute specified algorithms correctly (according to specified correctness criteria) in the presence of hardware and/or software faults; 2) the testing and verification of the initial correctness of hardware and software systems prior to utilization; 3) the design and implementation of on-line fault detection, fault location, and system reconfiguration procedures that can be used to recover from hardware and software faults, to perform system maintenance, and to maintain security; and 4) the development of models, measures, and techniques for evaluating the reliability, availability and, in general, the effectiveness of fault-tolerant computing systems.