The stochastic analysis of technical redundant systems.

Abstract

Today computers are widely used and their failures can lead to catastrophe. Redundancy is used to improve computer system reliability. This study presents a qualitative analysis of triple modular redundant (TMR) systems with redundant voters that can help to reduce the risk of failures. The voters examine the output of triple modular systems. Markov technique was employed to perform the analysis. Generalized formulas for reliability and mean time to failures of TMR systems with N voters have been developed. In addition, the system state probabilities in s-domain and steady state probabilities have been presented. The study clearly demonstrates improvements in system reliability, mean time to failure and steady state availability by the introduction of repair and redundancy in these systems.