IMPROVED CLUSTER MANAGEMENT METHOD FOR INDUSTRIAL “INTERNET OF THINGS” NETWORKS

Abstract

Background. With growing technological level and the corresponding availability of computational devices, sharply appears the problem of providing the high availability of high-loaded systems, as well as the reliability of the data storage and integrity. Therefore, it is not surprising that a large number of solutions have put high demands to the data manipulation.
Objective. The aim of the paper is to create a generic algorithm for industrial IoT networks that will be used for managing containerization components in database clusters.
Methods. During the work process most of existing solutions for providing automatic scalability of database cluster have been analyzed. Based on this analysis a special algorithm that manipulates the images of database cluster nodes based on predefined input parameters was proposed.
Results. Algorithm for managing database clusters in industrial IoT networks that will allow clients to optimize the resource usage with saving availability and incoming throughput characteristics of the cluster.
Conclusions. On the first stage, using this approach in production solutions requires more preparation and administration expenses, compared with existing solutions, but with time, it will give an advantage because the system will fully support the cluster to keep it in optimal condition without human interruption.