This chapter reviews the development of the recovery block approach to software fault
tolerance and subsequent work based on this approach. It starts with an account of the …

N-version programing is defined as the independent generation of N2 2 functionally
equivalent prograns fran the sme initial specification. A methodology of N-version …

Reliability is a desirable feature for any computing system, and a necessary requirement for
some systems, as discussed in Chapter 1. Although operation without failure is the goal, it …

We present a new approach, called local discriminant embedding (LDE), to manifold
learning and pattern classification. In our framework, the neighbor and class relations of data …

The problem of scheduling periodic time critical tasks on a monoprocessor computer system
is examined. Such tasks consist of an infinite number of requests, each with a prescribed …

Above all, it is vital to recognize that completely guaranteed behavior is impossible and that
there are inherent risks in relying on computer systems in critical environments. The …

An N-version software (NVS) unit is a fault tolerant software unit that depends on a generic
decision algorithm to determine a consensus result from the results delivered by two or more …

This book presents papers discussing the aspects of making computer applications systems
reliable-fault diagnosis and fault tolerance. It provides a current perspective of the testing …

It is essential to assess the reliability of digital computer systems used for critical real-time
control applications (eg, nuclear power plant safety control systems). This involves the …

Sender-based message logging is a new low-overhead mechanism for providing transpar-
ent fault-tolerance in distributed systems. It differs from conventional message logging …