Evolutionary Algorithm with AHP Decision-Making Method for Cloud Workflow Service Composition

Multiple Cloud services can be composed as a single service to fulfill the execution of large-scale workflow application. It faces trade-offs among several QoS metrics causing by users´ preferences. Most existing works utilize multi-objective evolutionary algorithms to address this problem. However, it is hardly to express different users´ preferences. This paper proposes an evolutionary algorithm for Cloud service composition, combining NSGA-II with Decision-Making method to calculate Crowding Distance with triangular fuzzy number based on AHP. The experimental results show that our approach could precisely capture users´ preferences, and perform better in scalability than other general multi-objective algorithms.