云计算环境下基于改进离散萤火虫群优化算法的服务选择方法研究

发布时间：2018-07-16 09:03

【摘要】：随着互联网和计算机技术的不断发展,云计算倡导一切皆服务,将云端丰富的计算资源、存储资源、平台服务、个性化应用等封装成云服务并通过基于Web的工具或者应用软件供给用户,这给传统的服务模式带来了巨大的挑战。由于云计算环境下的Web服务数量急剧增加,使用传统的服务选择方法已不能满足用户复杂且又动态的需求,如何利用云计算技术快速、高效地从海量的候选服务集中选择出服务质量较优的具体服务,从而形成满足用户需求的组合服务,已成为亟须解决的难题。为了解决在云计算环境下的大规模服务选择问题,论文应用了Web服务技术、离散萤火虫群优化算法(DGSO)以及云计算技术,并将离散萤火虫群优化算法融入到云计算技术中,实现在云计算环境下萤火虫并行计算。主要研究内容包括以下这两点:(1)为了提供传统选择方法对海量Web服务选择的支持,论文提出了基于改进离散萤火虫群优化算法(IDGSO)的服务选择方法。该算法重新定义了个体的编码、个体间的距离计算公式,改进了位置更新、非可行解处理和相关参数,并结合分群理论,提高了种群多样性和在高维空间中的搜索能力。(2)针对云计算环境下的海量Web服务,论文运用云计算的Map Reduce模式将改进离散萤火虫群优化算法并行化,提出了基于MRDGSO算法的服务选择方法。该算法重新定义了萤火虫的编码方式,并且融入分群分治策略,实现在云环境下小规模多种群并行计算。实验结果表明该算法在扩展性和性能方面均有良好的表现。
[Abstract]:With the continuous development of the Internet and computer technology, cloud computing advocates all services, will cloud rich computing resources, storage resources, platform services, Personalized applications are encapsulated into cloud services and supplied to users through Web-based tools or application software, which brings great challenges to traditional service modes. Because of the rapid increase of the number of Web services in cloud computing environment, the traditional service selection method can no longer meet the complex and dynamic needs of users. It has become an urgent problem to select the best quality of service from the mass candidate service set and to form a composite service to meet the needs of users. In order to solve the problem of large-scale service selection in cloud computing environment, the paper applies Web services technology, discrete firefly swarm optimization algorithm (DGSO) and cloud computing technology, and integrates discrete firefly swarm optimization algorithm into cloud computing technology. Implement the parallel computing of fireflies in cloud computing environment. The main research contents are as follows: (1) in order to provide the support of traditional selection method for mass Web service selection, a service selection method based on improved discrete Firefly swarm optimization algorithm (IDGSO) is proposed in this paper. The algorithm redefines the coding of individuals, the calculation formula of distance between individuals, improves the updating of position, the processing of infeasible solutions and the related parameters, and combines with the theory of clustering. Population diversity and searching ability in high-dimensional space are improved. (2) aiming at the massive Web services in cloud computing environment, the paper uses Map reduce mode of cloud computing to parallelize the improved discrete firefly swarm optimization algorithm. A service selection method based on MRDGSO algorithm is proposed. The algorithm redefines the encoding method of fireflies and integrates the clustering strategy to realize multi-cluster parallel computing in cloud environment. The experimental results show that the algorithm has good performance in expansibility and performance.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;TP393.09

【相似文献】