Ensuring the formation of control actions for computer node control on the basis of stochastic approximation projection algorithms

Бесплатный доступ

In practice, critical tasks require prompt management decisions using computer systems. Computer systems provide a flexible response to a variety of possible changes in the external environment, including, subject to an increase in the scope of information required for processing. Dynamic changes in the external environment necessitate rapid changes in industrial enterprises. These internal changes should be developed in concert with external ones, and, in most cases, they can be significant and abrupt. The active use of computer systems creates a lot of risks associated with cyber attacks. These risks are related to business processes based on the use of automated computer systems. The significance of the problem of managing computer nodes is actually obvious, since the implementation of the planned modifications of the computer infrastructure in practice allows to perform a variety of previously unattainable, fundamentally new steps in order to ensure the development of the infrastructure itself. Practical measures for computer infrastructure operation demonstrate that the constant provision of information and communication complexes and systems requires, first of all, a continuous analysis of the existing infrastructure and its subsequent adjustment. Therefore, within the framework of this publication, the provision of development measures offor computer node control based on projection algorithms of stochastic approximation is considered. The operation of computer nodes, taking into account their management using recurrent stochastic optimization algorithms, makes it possible to implement changes in the existing relationships in the computerized infrastructure field, which will provide development opportunity for organizations based on such infrastructures.

Еще

Computer nodes, information infrastructure, stochastic approximation, control actions, recurrent algorithms

Короткий адрес: https://sciup.org/140297113

IDR: 140297113   |   DOI: 10.18469/ikt.2022.20.2.07

Статья научная