Comparative Analysis of Container Build Methods: A Performance Evaluation

While containerization technology becomes a standard in software application development, challenges arise in the creation of efficient container images. Although Docker has historically dominated the containerization ecosystem, the process of crafting a Dockerfile for image creation can be complex and time-consuming, requiring meticulous attention to Docker’s established best practices. To address this challenge, various tools, including Buildpacks, Jib, and BuildKit, have been proposed. This paper presents a comprehensive experimental study evaluating these container build methods within the scope of a Docker-based containerized cluster. Performance evaluation tests were conducted to compare their effectiveness in terms of build duration, image size, and upload/download time. The results empower engineers to make informed decisions, enhancing efficiency and resource utilization in the development process for containerized applications.