Variability Modeling for Parameter-based Configurable Software Architectures
Kilian Bryon, Jens Vankeirsbilck, Jeroen Boydens · 2024
Configurable software architectures reuse common functionalities in numerous software products to reduce total development time and increase product quality. Testing these architectures can become a significant challenge due to the large number of possible configurations and limited testing time. Variability modeling can assist in making intelligent decisions on which combinations to test first, increasing various coverage goals. As opposed to using strict variants of a software module, such high configurability is often achieved by setting various parameters. This paper first studies the Universal Variability Language (UVL) and High-Level Variability Language (HLVL) to assess whether these are eligible for modeling these parameter-based configurable software architectures. Secondly, based on the outcome of this study, we introduce a new variability model language, called Parameter Variability Language (PVL), to better suit the needs for parameter-based variability.