THE STUDYING OF THE PROBABLE INDICATOR OF FUNCTIONAL STABILITY OF DISTRIBUTED INFORMATION SYSTEMS

Abstract

Currently, there is an active informatization of various spheres of society. This is the main reason why distributed information systems are increasingly being used. Since these information systems are often quite large, the need to develop and modify methods for assessing their functional stability is expected.

During the last decades, many indicators of functional stability of distributed information systems and methods of their calculation have been described. These include edge and vertex connectivity, probability and connectivity matrix, etc. However, the calculation of some of them involves a very large number of calculations and, sometimes, other equally complex accompanying processes. Therefore, there is a need to modify these indicators of functional stability in order to optimize their calculation.

This paper examines the modification of such an indicator of functional stability as the convolution of the connectivity matrix. Unlike other works, it is proposed to use a predefined kernel instead of one that depends on certain parameters. Due to this, it is possible to simplify the use of this probability indicator due to the absence of the need to periodically recalculate the kernel in case of changes in the system parameters. The peculiarities of the behavior of the described convolution in the case of the dependence of the connectivity matrix on the probability of serviceability of communication lines are also demonstrated. As a result of this, it is shown that with sufficiently large values of this parameter, its small fluctuations have almost no effect on the functional stability of the distributed information system, that is, it will still be close to the maximum possible.

The results obtained in the work can be used for a more optimal assessment of the functional stability of distributed information systems and the study of the influence of various parameters on it.