Successive Operational Periods as Measures of Dependability
We consider fault-tolerant computing systems, that is, systems which are able to recover an operational state after a fault. We propose new measures of dependability to quantify the behaviour of such a system all along its lifetime. With respect to classical measures (point availability, reliability) we consider the successive periods during which the system is in operation. Under markovian assumptions, we give closed-form expressions of the distribution and the moments of these operational periods. These measures give more insight on the evolution of the system than classical ones. Their utilization is illustrated by means of a numerical example.
KeywordsSojourn Time Operational Period Transition Probability Matrix Hardware Failure Local Clock
Unable to display preview. Download preview PDF.
- J.C.Laprie, “Dependable Computing and Fault Tolerance: Concepts and Terminology”, In 15th IEEE Int. Symp. on Fault Tolerant Computing (FTCS-15), Ann Arbor (Michigan), 1985.Google Scholar
- B.Parhami, “From Defects to Failures: a View of Dependable Computing”, Computer Architecture News, 16(4), September 1988.Google Scholar
- R. Marie, B. Sericola, “Distribution du temps total de séjour dans un sous-ensemble d’états transitoires d’un processus Markovien homogène à espace d’état fini”, Technical Report 585, I.N.R.I.A., Campus de Beaulieu, 35042 Rennes Cedex, France, November 1986.Google Scholar
- G. Ciardo, R. Marie, B. Sericola, K. Trivedi, “Performability analysis using semi-Markov process”, Technical Report CS-1988-9, Duke Computer Science, 1988, to appear in IEEE Trans. on Comp.Google Scholar
- G. Le Lann, “Algorithms for Distributed Data-Sharing Systems which use Tickets”, In Proc. 3rd Berkeley Workshop on Distributed Data Base and Computer Networks, Berkeley, USA, 1978.Google Scholar
- J. Misra, “Detecting Termination of Distributed Computations Using Markers”, In Proc. 2nd annual ACM Symposium on Principles of Distributed Computing, Montreal, Canada, 1983.Google Scholar
- M. Raynal, G. Rubino, “An Algorithm to Detect Token Loss on a Logical Ring and to Regenerate Lost Tokens”, In Proc. of the International Conference on Parallel Processing and Applications, L’Aquila, Italie, North-Holland, 1987.Google Scholar
- M. Raynal, G. Rubino, Détecter la perte de jetons et les régénérer sur une structure en anneau, Technical Report 428, I.N.R.I.A., Campus de Beaulieu, 35042 Rennes Cedex, France, Juillet 1985.Google Scholar