Exploring the Relationship between Dockerfile Quality and Project Characteristics

Dockerfile plays an important role in the Docker-based software development process, but many Dockerfile codes are infected with quality issues in practice. Previous empirical studies showed the existence of association between code quality and project characteristics. However, the relationship between Dockerfile quality and project characteristics has never been explored. In this paper, we seek to empirically study this relation through a large dataset of 6,334 projects. Using linear regression analysis, when controlled for various variables, we statistically identify and quantify the relationship between Dockerfile quality and project characteristics.