基于服务质量的Web服务优化选择算法及仿真

Quality of Servicebased Web Services Optimal Selection Algorithm and Its Simulation

摘要:

提出一种基于服务非功能属性的Web服务优化选择的多目标遗传算法.考虑到在服务选择时组合方案中的控制结构、具体服务之间的关联性和多个服务质量指标之间的折衷这三方面的约束,对服务选择建立分层模型给出优化选择的形式化定义.在此基础上,采用多目标遗传算法求解优化服务选择.首先设计了适合的染色体编码方式,以表示可行的服务选择方案,改进基本的遗传算子,采用两种遗传策略来避免算法陷入局部最优,以保证种群的收敛性.仿真结果表明,该方法能在有限进化代数内产生一组满足用户服务质量需求的非劣服务组合方案解,遗传策略在提高算法收敛性和维护种群多样性方面也发挥了作用.

Abstract:

The paper presents a quality of servicebased(QoS) Web services optimal selection method by multiobjective genetic algorithm by taking three aspects of constrains in services selection process into consideration, such as control structure within a composition plan, relationship between concrete services, and tradeoff among multiple QoS indexes. Moreover, a layered services selection model is constructed for formalizing optimal services selection problem,and a multiobjective genetic algorithm is first adopted. In particular,a chromosome coding method is first designed to represent a feasible services selection solution, and then genetic operators and strategies are developed to maintain diversity of population and also avoid getting trapped in local optima. Finally, experimental results show that this algorithm can generate a set of nondominated Pareto optimal solutions which satisfy to users’ QoS requirements within a finite number of evolving generation.The adopted genetic strategies have good effect on maintaining diversity of population and increasing convergence of algorithm.