Performance Analysis of Docker-based Database Management Systems Compared to Virtual Machine-based Systems: A Comparative Study

Abstract

Computer virtualization is a very old technology. Due to a lot of technical barriers, computer containerization has been introduced recently. Nowadays, computer containerization is playing a major role in information technology and containerization is a trending topic. Among the practitioner of information technology, a lot of software services are moving to containerization instead of traditional virtual machines. Among the most famous software services: database management systems are a leading service. Among most computer containerization technologies, Docker is the most popular and trending container vendor. Therefore, identification of the performance of database management systems on the Docker-based platform is an essential task for the practitioner. This research study aims to identify the practical aspects of each database management system on the Docker-based infrastructure for main database management system operations. For the study: Ubuntu 18.04 Long Term Support (used package with architecture: GNU/Linux 4.15.0-112-generic x86_64) cloud-based operating system was used and on that operating system the Docker infrastructure was launched. Docker version 19.03.9 was launched for the study. On Docker: MySQL, PostgreSQL, and MongoDB database management system containers were launched separately. SELECT, DELETE, UPDATE, and INSERT operations were used for the performance evaluations of database management system response times. This research identified that there was an increase in the performance of the Docker platform with a 95% confidence interval level for all data records to virtual machine-based platforms. Finally, the research study identified that Docker-based database management system has a quick response time than virtual machine-based database management systems.