A Reproducible Tutorial on Reproducibility in Database Systems Research

Reproducibility is a key aspect of the scientific method, and it is essential for building trust in the results of research. This tutorial aims to provide concrete guidance on how to leverage containerized reproducibility using Docker for database systems research. In this tutorial, we present a step-by-step guide on how to prepare a Docker-based artifact for an experiment. We will cover topics such as Dockerfiles, Docker images, Docker Compose, automation using Python, Bash, and Make, and also artifact documentation and packaging best practices. The tutorial itself is a reproducible artifact, and we provide a public GitHub repository with all the code and examples used in the tutorial. This repository can serve as a starting point to prepare artifacts for experiments and publications.