A fuzzy approach for ranking adaptation strategies in CLAM

Selecting an appropriate adaptation strategy in service-based systems (SBS) is a complex issue. Basically a problem in the system can be addressed using various sets of adaptation actions where each one has different consequences. In this paper we propose a new adaptation selection approach for a cross-layer adaptation manager (CLAM), which makes an exhaustive impact analysis to discover alternative adaptation strategies for SBS. Each alternative strategy is a series of adaptation actions that might emerge from different parts of the SBS. Our approach uses a ranker module based on fuzzy logic and the selection is carried out by inferring various criteria using a fuzzy inference system (FIS).