МОДЕЛЬ ФОРМИРОВАНИЯ КЛАСТЕРОВ ИНФОРМАТИВНЫХ УЗЛОВ ИНТЕГРИРОВАННОЙ И РАСПРЕДЕЛЕННОЙ ОБРАБОТКИ ДАННЫХ В ВЫЧИСЛИТЕЛЬНОЙ СЕТИ

Abstract

The spread of network and computational technologies is relevant when solving applied issues in various areas of human activity. One of these issues is the accumulation of data in network and computing structures. The solution of the corresponding problem is made on the basis of integrated and distributed data processing in computer networks, which is characterized by the use of intelligent and cognitive data analysis, as well as distance metrics. The purpose of this study is to develop a mathematical model that takes into account the peculiarities of the computing activity of network node devices to calculate its informative load for the estimated time and the formation of clusters of informative nodes (their spaces) in computer networks. On the basis of data mining, it is proposed to consider deviations in the computing activity of network devices as information content of a network node, the ability to generate big data. In further modeling, the methods of the theories of pattern recognition and simulation were used, as well as the distance metrics toolkit, which allow to formalize heterogeneous data, taking into account the peculiarities of computer networks. An element of novelty of the proposed mathematical model is that it takes into account the peculiarities of the data of the computing activity of network node devices for the estimated time by using the Manhattan metric. A computational example of the formation of clusters of informative network nodes and their spaces is given. In the calculations, the author's sensor for modeling random variables is used, which significantly increases the quality of calculations and their clarity. The practical significance of the proposed mathematical model lies, first of all, in the fact that it becomes possible to track the dynamics of changes in information content in network structures with a large number oencf nodes. This makes it possible to control data processing, in addition, to increase the efficiy of computer networks by taking preventive measures to maintain their technical condition and information security.