A Partial Selection Methodology for Efficient QoS-Aware Service Composition
Ying Chen,Jiwei Huang,Chuang Lin,Jie Hu
DOI: https://doi.org/10.1109/tsc.2014.2381493
IF: 11.019
2014-01-01
IEEE Transactions on Services Computing
Abstract:As web service has become a popular way for engineering software on the Internet, quality of service (QoS) which describes non-functional characteristics of web services is often employed in service composition. Since QoS is an aggregated concept consisting of several attributes, service composition on enormous candidate sets is a challenging multi-objective optimization problem. In this paper, we study the problem from a general Pareto optimal angle, seeking to reduce search space in service composition. Pareto set model for QoS-aware service composition is introduced, and its relationship with the widely used utility function model is theoretically studied, which proves the applicability of our model. QoS attributes are systematically studied according to their different types of aggregation patterns in service composition, and QoS-based dominance relationships between candidates and between workflows are defined. Taking advantage of pruning candidates by dominance relationships and constraint validations at candidate level, a service composition algorithm using partial selection techniques is proposed. Furthermore, a parallel approach is designed, which is able to significantly reduce search space and achieve great performance gains. A careful analysis of the optimality of our approach is provided, and its efficacy is further validated by both simulation experiments and real-world data based evaluations.