【摘要】 针对云计算环境下存在虚拟化资源利用率不高、延迟、性能衰减等问题,提出一种基于粒子群的虚拟资源分配优化方法。根据云计算优势,对虚拟化资源进行描述及状态定义,分析了虚拟环境下的服务质量(quality of services,QoS)组成和计算方法;同时通过用户与云提供商收益最大化的平衡关系,建立一种虚拟化操作方法,以确保虚拟化资源被合理分配和回收;采用粒子群算法使用户和虚拟服务提供方都能趋于收益平衡。在此基础上设计的一款云播放器与直接嵌入到Web播放器相比得出,提出的粒子群优化方法有效且将资源利用率至少提高10%,与QoS约束的二元整数规划方法相比,提出的方法操作虚拟资源更有效。更多还原

【Abstract】 In order to solve virtual problems of cloud computing such as efficiency,delay and attenuation performance. An approach based on PSO( particle swarm optimization) is presented in the paper. According to the advantage of cloud computing,some properties of virtual resources( virtual machine: VM) are described and defined,and QoSs( quality of services) and computing methods of VM are also defined and set; after that balance relation and maximum profits of cloud users( U) and VM providers,a VM’s operate approach is described,which is sure to be distributed and retrieved the VM; And we employ and design PSO to accomplish profits balance optimization for U and VM on the basis of the research results,a cloud player is desgined and realized,compare with a web player,the approach show that is effective,and the result improved at least 10%. Meanwhile,PSO-based approach is more effective than BIP( binary interger programming)-based QoS constrained allocation.更多还原