Software Fault Tolerance A Tutorial

Because of our present inability to produce error-free software, software fault tolerance is and will continue to be an important consideration in software systems. The root cause of software design errors is the complexity of the systems. Compounding the problems in building correct software is the difficulty in assessing the correctness of software for highly complex systems. After a brief overview of the software development processes, we note how hard-to-detect design faults are likely to be introduced during development and how software faults tend to be state-dependent and activated by particular input sequences. Although component reliability is an important quality measure for system level analysis, software reliability is hard to characterize and the use of post-verification reliability estimates remains a controversial issue. For some applications software safety is more important than reliability, and fault tolerance techniques used in those applications are aimed at preventing catastrophes. Single version software fault tolerance techniques discussed include system structuring and closure, atomic actions, inline fault detection, exception handling, and others. Multiversion techniques are based on the assumption that software built differently should fail differently and thus, if one of the redundant versions fails, it is expected that at least one of the other versions will provide an acceptable output. Recovery blocks, N-version programming, and other multiversion techniques are reviewed.Torres-Pomales, WilfredoLangley Research CenterSOFTWARE ENGINEERING; SOFTWARE RELIABILITY; COMPUTER PROGRAMMING; COMPUTER PROGRAMS; FAULT TOLERANCE; APPLICATIONS PROGRAMS (COMPUTERS); SOFTWARE DEVELOPMENT TOOLS; OPERATING SYSTEMS (COMPUTERS); COMPUTER SYSTEMS DESIGN; COMPLEX SYSTEMS; COMPONENT RELIABILITY; QUALITY CONTROL

The first ESPRIT programme contained several ambitious projects. of which REQUEST. with its wide brief covering all issues of assessment of quality and reliability of software process and product. was one. Within REQUEST. the research described in this volume. concerning those special problems of software that is required to have extremely high reliability. was particularly difficult and ambitious. The problems of software reliability are essentially twofold. On the one hand there is a concern with methods for achieving adequate reliability. on the other hand there is a need to evaluate what has actually been achieved in a particular case. Naturally. far more effort has been spent over the years on the former problem; indeed. there is a sense in which all of conventional software engineering can be seen as a response to this problem. However. it is becoming clearer than ever that we can only claim to have a truly sCientific approach. and so justify the description software engineering. when we are able to measure the attributes of process and product. It is still common to find software development methods recommended to users on purely anecdotal grounds. This is not good enough. Rational choices between rival approaches can only be made on the basis of quantified costs and benefits. Even more worrying is the tendency to argue that a software product can be depended upon merely because it has been developed by honest men using such anecdotal 'good practice'.

This innovative resource provides the most-comprehensive coverage of software fault tolerance techniques as it guides professionals through their design, operation and performance. It features an in-depth discussion on the advantages and disadvantages of specific techniques, so practitioners can decide which ones are best suited for their work.

Software fault tolerance techniques involve error detection, exception handling, monitoring mechanisms, and error recovery. This issue of Trends in Software focuses on identification, formulation, application, and evaluation of current software fault tolerance techniques.

In architecting dependable systems, what is required to improve the overall system robustness is fault tolerance. Many methods have been proposed to this end, the solutions are usually considered late during the design and implementation phases of the software life-cycle (e.g., Java and Windows NT exception handling), thus reducing the effectiveness error and fault handling. Since the system design typically models only normal behaviour of the system while ignoring exceptional ones, the implementation of the system is unable to handle abnormal events. Consequently, the system may fail in unexpected ways due to faults.It has been argued that fault tolerance management during the entire life-cycle improves the overall system robustness and that different classes of threats need to be identified for and dealt with at each distinct phase of software development, depending on the abstraction level of the software system being modelled.This book builds on this trend and investigates how fault tolerance mechanisms can be applied when engineering a software system. In particular, it identifies the new problems arising in this area, introduces the new models to be applied at different abstraction levels, defines methodologies for model-driven engineering of such systems and outlines the new technologies and validation and verification environments supporting this.

Software patterns have revolutionized the way developer’s and architects think about how software is designed, built and documented. This new title in Wiley’s prestigious Series in Software Design Patterns presents proven techniques to achieve patterns for fault tolerant software. This is a key reference for experts seeking to select a technique appropriate for a given system. Readers are guided from concepts and terminology, through common principles and methods, to advanced techniques and practices in the development of software systems. References will provide access points to the key literature, including descriptions of exemplar applications of each technique. Organized into a collection of software techniques, specific techniques can be easily found with sufficient detail to allow appropriate choices for the system being designed.