A Novel Hybrid Bio-Inspiration Technique for Service Composition

Abstract

There are many challenges facing the service composition process. These challenges include, how to integrate services to satisfy global user requirements, missing or changeable values of QoS, and how to reduce the large solution space of candidate services. In this paper, we proposed a framework to address these challenges. The proposed framework consists of three phases. The Normalizer phase gives a certain range for all QoS attributes and historical user orders. During the Clustering phase, the search space is reduced. Finally, the composition process is done, and a list of candidate composite services is generated through the Service composer phase. We present the hybrid bio-inspiration technique to implement the proposed framework and prove its applicability. In addition, we introduced the MR-FPSO algorithm to implement this phase by merging PSO and FOA optimization algorithms over the MapReduce framework to handle the large scale of data in the cloud environment. Our technique is compared to different techniques, including MR-GA, MR-IDPSO, and MRPSO. The simulation results proved that our technique outperforms the other techniques.