ОСОБЛИВОСТІ МАСШТАБУВАННЯ КОНТЕЙНЕРНОГО НАВАНТАЖЕННЯ НА БАЗІ СИСТЕМИ KUBERNETES

Abstract

Today, the field of software development using container technology is becoming widespread in many areas of business, and the problem of scaling is one of the most important to achieve the expansion of the volume of tasks performed over time, so it is important to use the right scaling schemes. Kubernetes requires relatively easy-to-adapt container management mechanisms, but the problem of scaling them remains poorly described and researched, which increases the complexity of the business transition to container technologies and, in particular, the Kubernetes system. Thanks to the separation of infrastructure and payload concepts in Kubernetes, the classic scaling methods – horizontal and vertical – are revealed in each component separately. Although the Kubernetes system is quite flexible, it has only recently been developed, and the features of scaling with concept separation have not yet been described. Actual scientific researches and issues analysis showed that there is a general trend towards cloud migration recently. In this context, the architectural style of microservices and the use of container load has attracted considerable attention, especially based on the Kubernetes system. It can be concluded that Kubernetes has significant advantages in the speed of development and deployment of container-based applications, but the problem of scaling and features of horizontal and vertical scaling in Kubernetes has not received much attention from academia so far. The aim of the work is to study the features of scaling the container load based on the Kubernetes system. To explain the basic operations to achieve horizontal and vertical scaling. To describe the scalability property of the Kubernetes system. This work reveals the main architectural features of Kubernetes, the principles of working with applications in container virtualization. The main methods of regulating the resource capacity of containers are shown. The concepts of horizontal and vertical scaling in the Kubernetes system with an indication of scalability properties are revealed. The main operations for scaling in the Kubernetes system in the context of separation of payload and infrastructure concepts are given. In this article the features of container load scaling based on the Kubernetes system with a separation of the concepts of payload and infrastructure scaling were investigated and described. This article is a review.