QoS-aware Big service composition using MapReduce based evolutionary algorithm with guided mutation

Big services are the collection of interrelated services across virtual and physical domains for analyzing and processing big data. Big service composition is a strategy of aggregating these big services from various domains that addresses the requirements of a customer. Generally, a composite service is created from a repository of services where individual services are selected based on their optimal values of Quality of Service (QoS) attributes distinct to each service composition. However, the problem of producing a service composition with an optimal QoS value that satisfies the requirements of a customer is a complex and challenging issue, especially in a Big service environment. In this paper, we propose a novel MapReduce-based Evolutionary Algorithm with Guided Mutation that leads to an efficient composition of Big services with better performance and execution time. Further, the method includes a MapReduce-skyline operator that improves the quality of results and the process of convergence. By performing T-test and Wilcoxon signed rank test at 1% level of significance, we observed that our proposed method outperforms other methods.