Data-intensive Service Mashup based on Game theory and Hybrid Fireworks Optimization Algorithm in the Cloud

摘要

End users can create kinds of mashups which combine various data-intensive services to form new services. The challenging issue of data-intensive service mashup is how to find service from a great deal of candidate services while satisfying SLAs. In this paper, Service-Level Agreement (SLA) consists of two parts, which are SLA-Q and SLA-T. SLA-Q (SLA-T) indicates the end-to-end QoS (transactional) requirements. SLA-aware service mashup problem is known as NP-hard, which takes a significant amount of time to find optimal solutions. The service correlation also exists in data-intensive service mashup problem. In this paper, the service correlation includes the functional correlation and QoS correlation. For efficiently solving the dataintensive service mashup problem with service correlation, we propose an approach GTHFOA-DSMSC (Dataintensive Service Mashup with Service Correlation based on Game Theory and Hybrid Fireworks Optimization Algorithm) which evolves a set of solutions to the Pareto optimal front. The experimental tests demonstrate the effectiveness of the algorithm.