Fault tolerance via diversity against design faults (tutorial session): design principles and reliability assessment

ABSTRACT

Research results indicate that (as usual in software engineering) these question can only be answered with reference to each specific application context and that diversity is no “silver bullet”. But diversity is an attractive option, made more interesting by current trends like the preference for COTS items, and it is important for practitioners to go beyond the summary opinions and misunderstanding that surround it.This tutorial is designed for people involved in system design, acceptance or certification, especially in companies with high dependability requirements or plans to improve on current levels to move into more demanding markets. It is also appropriate for researchers in software engineering wishing to obtain an up-to-date view of knowledge in this area.This tutorial describes:the motivations behind the use of software fault tolerance, and thus the circumstances in which it should be considered as a possible choice;what design schemes one may adopt, and which issues a designer needs to be aware of, for effective application. We present both examples of industrial use and explanations of the important design choices and trade-offs. In this part, we cover the widely published solutions of N-version programming and recovery blocks, but also describe the various options available to a designer, and interesting specific solutions adopted in the railway and aviation industry, and scheme for applications to safety systems. We discuss the factors that may decide the scheme to be adopted and the design of adjudication between conflicting results; “what one should really believe” about the effectiveness of software fault tolerance in improving reliability, beyond the controversy and the misunderstandings surrounding it. We give a picture, assembled from more than 10 years of research, of what evidence has really been produced for and against software diversity. We explain the weaknesses of the extreme opinions voiced for and against software fault tolerance, and discuss the criteria that should affect practical decisions about using it, about how to improve its effectiveness by appropriate decisions in developing alternate versions of software components, and about its value for system acceptance.

INDEX TERMS

Primary Classification:
  D. Software
  D.2 SOFTWARE ENGINEERING

Additional Classification:
  A. General Literature

  D. Software
  D.2 SOFTWARE ENGINEERING

General Terms:
Design, Performance, Reliability, Theory