Acapulco: an extensible tool for identifying optimal and consistent feature model configurations

Abstract

Configuring feature-oriented variability-rich systems is complex because of the large number of features and, potentially, the lack of visibility of the implications on quality attributes when selecting certain features. We present Acapulco as an alternative to the existing tools for automating the configuration process with a focus on mono- and multi-criteria optimization. The soundness of the tool has been proven in a previous publication comparing it to SATIBEA and MODAGAME. The main advantage was obtained through consistency-preserving configuration operators (CPCOs) that guarantee the validity of the configurations during the IBEA genetic algorithm evolution process. We present a new version of Acapulco built on top of FeatureIDE, extensible through the easy integration of objective functions, providing pre-defined reusable objectives, and being able to handle complex feature model constraints.