Cloud Service Selection Based on Variability Modeling

The selection among Cloud services is a recent problem in research and practice. The diversity of decision-relevant criteria, configurability of Cloud services and the need to involve human decision-makers require holistic support through models, methodologies and tools. Existing Cloud service selection approaches do not address all stated difficulties at the same time. We present an approach to capture capabilities of Cloud services and requirements using variability modeling. We use Cloud feature models (CFMs) as a representation mechanism and describe how they are utilized for requirements elicitation and filtering within a presented Cloud service selection process (CSSP) that includes human decision-makers. Filtering produces a reduced number of valid Cloud service configurations that can be further assessed with current multi-criteria decision making-based selection approaches. We present software tools that we use to demonstrate the applicability of our approach in a use case about selecting among Cloud storage services.