基于QoS的云服务组合技术研究

发布时间：2018-01-29 07:46

本文关键词： 云服务组合 QoS属性模型组合算法　出处：《大连海事大学》2015年硕士论文　论文类型：学位论文

【摘要】：随着云计算相关技术的迅速发展,单个云服务不能很好的满足用户多样化的需求,因此云服务组合问题得到广泛的关注和研究。在云服务组合中,以往的大多数研究没有考虑负载均衡等因素,可能导致组合方案在实际的运行中出现问题甚至无法执行,进而影响了云服务组合的质量和效率。首先,本文在阅读大量关于Web服务组合、云服务组合以及组合算法等相关文献的基础上,讨论了云环境中负载均衡等因素对云服务QoS的影响,并结合当前服务通用的QoS属性,给出了云服务QoS模型；同时,考虑云服务的开发平台对其QoS属性值的影响,给出了一种改进的云服务QoS属性的计算方法；进而给出了考虑全局最优的云服务组合的QoS属性计算模型。其次,采用遗传算法对本文研究的基于QOS的云服务组合问题进行求解,为改善遗传算法的缺陷,提出一种增强型的初始种群的生成方法,将统计关联上下文库选出的若干个个体和采用比例的方法生成的个体组合成一个较优的初始种群；此外,本文在遗传算法的选择操作以及交叉操作等方面也进行了改进。最后,编程实现了改进的遗传算法,并在matlab环境中进行仿真实验。通过实验数据的对比分析,验证了本文给出的基于QoS的云服务组合方法的合理性以及有效性。
[Abstract]:With the rapid development of cloud computing related technologies, single cloud services can not meet the diverse needs of users, so the cloud service composition problem has been widely concerned and studied. Most of the previous studies did not consider load balancing and other factors, which may lead to the composition scheme in the actual operation problems or even can not be implemented, and then affect the quality and efficiency of cloud service composition. First of all. On the basis of reading a lot of literatures about Web service composition, cloud service composition and composition algorithm, this paper discusses the influence of load balancing in cloud environment on cloud service QoS. Combined with the QoS attribute of current service, the QoS model of cloud service is given. At the same time, considering the influence of cloud service development platform on the value of QoS attribute, an improved computing method of cloud service QoS attribute is presented. Then the QoS attribute calculation model considering the global optimal cloud service composition is given. Secondly, the genetic algorithm is used to solve the cloud service composition problem based on QOS which is studied in this paper. In order to improve the defects of genetic algorithm, an enhanced initial population generation method is proposed. Several individuals selected from the statistical association context database and individuals generated by the proportional method are combined into an optimal initial population. In addition, the selection operation and crossover operation of genetic algorithm are also improved. Finally, the improved genetic algorithm is implemented by programming. Through the comparison and analysis of the experimental data, the rationality and validity of the cloud service composition method based on QoS are verified.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP393.09

【参考文献】