Comparative Analysis of Virtualization Methods in Big Data Processing

Abstract

Cloud computing systems have become widely used for Big Data processing, providing access to a wide variety of computing resources and a greater distribution between multi-clouds. This trend has been strengthened by the rapid development of the Internet of Things (IoT) concept. Virtualization via virtual machines and containers is a traditional way of organization of cloud computing infrastructure. Containerization technology provides a lightweight virtual runtime environment. In addition to the advantages of traditional virtual machines in terms of size and flexibility, containers are particularly important for integration tasks for PaaS solutions, such as application packaging and service orchestration. In this paper, we overview the current state-of-the-art of virtualization and containerization approaches and technologies in the context of Big Data tasks solution. We present the results of studies which compare the efficiency of containerization and virtualization technologies to solve Big Data problems. We also analyze containerized and virtualized services collaboration solutions to support automation of the deployment and execution of Big Data applications in the cloud infrastructure.