Dependable Computing and Fault Tolerance Research Group, LAAS-CNRS (France)

LAAS was established in 1968. It is a laboratory of the CNRS, the French National Organisation for Academic Scientific Research. LAAS has a permanent staff of over 270 (179 researchers and 95 supporting personnel, either technical or administrative), plus 150 doctorate students and 24 post-doc’. The permanent research-staff includes CNRS personnel (77), and faculty members (82), from University Paul Sabatier, the Toulouse Polytechnic Institute and the National Institute for Applied Sciences, which are three institutions officially associated with LAAS. Research carried out at LAAS covers Automatic Control, Computer Science and Engineering, Microelectronics, Robotics.

The research group on Dependable Computing and Fault Tolerance, which is one of twelve research groups at LAAS, has been active in the field of dependable computing systems for twenty four years. The group is composed of 11 senior researchers (CNRS and faculty members) plus 10 doctorate students. Its activities cover fault tolerance, fault removal and fault forecasting. The research covers a wide spectrum of fault classes and is structured according to four themes: i) fault tolerance in distributed systems, ii) software testing, iii) fault injection, and iv) probabilistic and statistical evaluation.

The overall aim of the group work is to develop methods for designing and validating dependable computing systems, together with their supporting tools. Our methods and tools are put into practice in applications from areas such as transportation, energy production, services, defence, space, etc. The formulation of the basic concepts of dependability is an underlying common factor for the group's activities. The definition of our methods is carried out within co-operative projects. In addition to the two IST projects that are currently running (DSoS and MAFTIA), four projects have been performed: i) Delta-4 (Definition and Design of an open Dependable Distributed systems architecture) in pre-competitive research, 1986-92, ii) PDCS (Predictably Dependable Computing Systems) in basic research, 1989-94, iii) DeVa (Design for validation), in Research Long Term, 1995-98, and iv) GUARDS (Generic Upgradable Architectures for Real-Time Dependable Systems), in Research Development, 1996-99.

With respect to the theme of the current proposal, it is noteworthy that the work performed on dependability evaluation and fault injection gave rise to the following six tools: SURF-2, for evaluation of dependability measures using stochastic Petri Nets and Markov chains, MEFISTO-L, a fault-injection system for VHDL simulation models, MAFALDA, a tool for analysing and implementing dependable operating systems based on commercial micro-kernels, SESAME, a software mutation environment aimed at validating software testing methods, SoRel, for the analysis and evaluation of software reliability from statistical failure data and MESSALINE a pin-level fault injector.

The group on “Dependable Computing and Fault Tolerance”, has established the Laboratory for Dependability Engineering (Laboratoire d’Ingénierie de la Sûreté de fonctionnement — LIS) together with five leading industrial companies: Aerospatiale Matra Airbus, Électricité de France, Matra Marconi Space France, Technicatome and Thomson-CSF. Hosted by LAAS from July 1992 to June 2000, LIS was a cooperative laboratory in which competencies and resources were put together to solve scientific and technical problems that were considered as key problems by all the partners. The results of these research activities could then be exploited by the various partners, according to the specialisation required by their respective application domains. It is worth mentioning the publication of “The Dependability Handbook” (in French) in 1995-96; also a first draft of the English version is ready and of “Software Components and Dependability — Integration of COTS” (in French) in 2000.

Relevant publications

J. Arlat, A. Costes, Y. Crouzet, J.-C. Laprie and D. Powell, “Fault Injection and Dependability Evaluation of Fault-Tolerant Systems,” IEEE Transactions on Computers, Vol. 42, No. 8, pp. 913-923, 1993.
J.-C Laprie and K. Kanoun, “Software Reliability and System Reliability”, Handbook for Software Reliability Engineering, Chapter 2, Ed. M. Lyu, McGraw-Hill, 1996.
J.C. Fabre, F. Salles, M. Rodriguez-Moreno, J. Arlat, “Assessment of COTS Microkernels by Fault Injection”, Proc. IFIP 7th Working Conf. on Dependable Computing in Critical Applications (DCCA-7), San Jose, CA, USA, pp. 19-38, January 1999 (IEEE CS Press).
K. Kanoun, M. Borrel, T. Morteveille and A. Peytavin “Availability of CAUTRA, a subset of the French Air Traffic Control System”, IEEE Transactions on Computers, Vol. 48, No. 5, pp. 528-535, 1999.
M. Rodríguez, F. Salles, J.-C. Fabre and J. Arlat, "MAFALDA: Microkernel Assessment by Fault Injection and Design Aid", in Proc. 3rd European Dependable Computing Conf. (EDCC-3), Prague, Czech Republic, pp. 145-160, September 1999 (Springer).

CVs of Key persons to be involved

Jean Arlat is Directeur de Recherche at LAAS-CNRS. His research interests focus on the evaluation of hardware-and-software fault-tolerant systems including both analytical modelling and experimental fault injection approaches, subjects on which he authored or co-authored more than seventy papers. From January 1997 to June 2000, he led the Laboratory for Dependability Engineering (LIS) that was a co-operative laboratory hosted by LAAS. He also conducted various National and European research contracts in ESPRIT and BRITE-EURAM Programmes, and has been a consultant to several companies in France and in Europe. He co-chaired the Program Committee of the 28th IEEE International Symposium on Fault-Tolerant Computing (FTCS-28). He is member of the IEEE Computer Society’s Technical Committee on Fault-Tolerant Computing that he chaired in 1994-1995. Since January 1999, he serves as Chair of the IFIP Working Group 10.4 on Dependable Computing and Fault Tolerance.

Jean-Charles Fabre received his Doctorate in Computer Science in 1982 from the University of Toulouse (France) and was first involved in the Chorus project at INRIA. Since 1984, he has been with the LAAS-CNRS in Toulouse working in the Dependable Computing and Fault Tolerance research group. His past and current interests concern distributed algorithms, implementation validation by fault-injection, fault and intrusion-tolerance in distributed systems. Today, his activity is concerned with object-oriented development of fault and intrusion dependable systems and validation of COTS microkernels by fault-injection. Author or co-author of more 40 publications in conference proceedings, journals and books, he has been involved in several ESPRIT projects (e.g. Delta-4, PDCS and DeVa) and is consultant for industrial companies. Dr. Fabre served in many program committees of international conferences and was program co-chair of the IEEE 1999 Symposium on Reliable Distributed Systems. He is member of the IEEE CS.

Karama Kanoun is Directeur de Recherche at LAAS-CNRS. She was Visiting Professor at the Univ. of Illinois, Urbana Champaign, the second semester of 1998. Her research interests include modelling and evaluation of computer systems. She has authored and co-authored more than ninety papers, conducted several research contracts. She has been involved in the ESPRIT projects: Delta-4, PDCS, Deva and GUARDS. She has been consultant for several European and French companies. She served as Program Committee (PC) co-chair of the International Symposium on Software Reliability Engineering (ISSRE'94), General Chair of ISSRE'95 and Safecomp'99, and PC co-chair of the international Conference on Dependable Systems and Networks in 2000. She was Guest Editor of a Special Issue on Dependability of Computing Systems, IEEE Transactions on Computers, in 1998. She is Associated Editor of IEEE Transactions on Reliability. She is a project reviewer for the EC and member of the IEEE Computer Society, and IEEE Technical Council on Software Engineering.

Jean-Claude Laprie is Directeur de Recherche of CNRS, the French National Organisation for Scientific Research. He joined LAAS-CNRS in 1968, where he has directed the research group on Fault Tolerance and Dependable Computing from 1975 to 1996. He has been Director of LAAS since 1997. His research has focused on dependable computing since 1973, and especially on fault tolerance and on dependability evaluations, subjects on which he has authored and co-authored more than 100 papers, as well as co-authored or edited several books. He also founded and directed, from 1992 to 1996, LIS, the Laboratory for Dependability Engineering, a joint academia-industry laboratory. In 1984-1985, he was Chairman of the IEEE Computer Society Technical Committee on Fault-Tolerant Computing, and has been Chairman of IFIP WG 10.4 on Dependable Computing and Fault Tolerance from 1986 to 1995. He received in 1992 the IFIP Silver Core, and in 1993 the Silver Medal of the French Scientific Research.