Dependability of Computer Systems
Objectives
Introduction to the basic concepts, assumptions and techniques for the design, the implementation and the evaluation of dependable computing systems, in general, and fault tolerant systems in particular. Illustration using several examples of dependable systems and experimental evaluation results.
Description
The lecture is composed of five section:
– Introduction, definitions and basic notions (fault prevention, fault tolerance, fault removal, fault forecasting) and measures.
– Fault tolerance techniques (fault assumptions, basic techniques, replication strategies) and architectural solutions
– Validation techniques, in particular, by fault injection (principles, robustness analysis, examples of tools and experimental results)
– Examples of dependable systems (real-time micro-kernel based systems, A320, B777, ELEKTRA)
– Software testing and verification/validation
Targeted skills
Knowledge of the development process of safety critical systems
Mastering assumptions and fault tolerant computing techniques
Characterisation techniques by fault injection and analysis
Bibliography
ENSEEIHT lecture slides (in English) presented by M. Fabre;
"Basic Concepts and Taxonomy of Dependable and Secure Computing"
Algirdas Avizienis, Jean-Claude Laprie, Brian Randell, and Carl Landwehr,
IEEE Transactions on Dependable ad Secure Computing, Vol. 1, No. 1, Jan. 2004.
Pre-requisites
Algorithmics, operating systems principle, real-time computing, C/C++ programming, computer architecture
Contact(s)
Places
- Toulouse