Mathematical Models for Determining the Reliability Characteristics of Distributed Information Processing Systems in the Electric Power Industry

Abstract

In this paper, using the exponential Queuing networks (QN) apparatus, the problem of finding the maximum fault tolerance distributed information processing systems (IPS) implemented on the basis of a two-level and three-level client-server architecture is formulated and solved. The method of finding the average response time for all user requests is presented. It takes into account the impact of fault tolerance of individual servers of such a system. The problem determining the fault tolerance indicators of distributed IPS was reduced to the problem determining the stationary probabilities of QN states. The paper also presents an iterative algorithm for solving this problem. At the conceptual level, the presented models can be used to obtain reliability characteristics of integrated automated control systems (IACS) in the electric power industry