Fuzzy logic based qos optimization mechanism for service composition

Increase emphasis on Quality of Service and highly changing environments make management of composite services a time consuming and complicated task. Adaptation approaches aim to mitigate the management problem by adjusting composite services to the environment conditions, maintaining functional and quality levels, and reducing human intervention. This paper presents an adaptation approach that implements self-optimization based on fuzzy logic. The proposed optimization model performs service selection based on the analysis of historical and real QoS data, gathered at different stages during the execution of composite services. The use of fuzzy inference systems enables the evaluation of the measured QoS values, helps deciding whether adaptation is needed or not, and how to perform service selection. Experimental results show significant improvements in the global QoS of the use case scenario, providing reductions up to 20.5% in response time, 33.4% in cost and 31.2% in energy consumption.