This is the replication package for the paper intitled:

"What Constitutes the Deployment and Run-time Configuration System? An Empirical Study on OpenStack Projects".

This paper investigates the configuration file types that constitute the configuration system of OpenStack. We conduct our study on OpenStack as it is one of the most deployed cloud platform for infrastructure management. To identify the different configruation files, we first manually investigate the different configuration types, where we identified 9 different types of configuration files. Then, we comapre between five classifiers (SVC, RF, LR, KNN, and GB), and leverage two machine learning models to automate the identification of the different configruation file types. The first model predicts the configuration from the non-configuration files. The second model predicts the different types of the configuration files.

In this package, we provide:

1. the datasets of training the first model and the second model. We also provide the links to the files of the first model and the second model second model
2. the scripts for building our first model and second model. Besides, we provide an R script to compute features redundancy
3. the comparison results between the five classifiers for the first model and the second model.

Please, use the following bibtex entry:

@article{bessghaier2023constitutes,
  title={What Constitutes the Deployment and Runtime Configuration System? An Empirical Study on OpenStack Projects},
  author={Bessghaier, Narjes and Sayagh, Mohammed and Ouni, Ali and Mkaouer, Mohamed Wiem},
  journal={ACM Transactions on Software Engineering and Methodology},
  volume={33},
  number={1},
  pages={1--37},
  year={2023},
  publisher={ACM New York, NY}
}