面向多通道QoS需求的虚拟机分簇挖掘仿真  

Multi-channel QoS Demand Oriented Virtual Machine Clustering Mining Simulation

作　　者：杨丽萍[1,2] 臧华中[1] 

机构地区：[1]三江学院计算机科学与工程学院,南京210012 [2]南京航空航天大学计算机科学与技术学院,南京210016

出　　处：《科技通报》2015年第12期206-208,共3页Bulletin of Science and Technology

摘　　要：实行虚拟机最优分配方案的准确挖掘能够提高云计算的服务质量(Qo S)。利用传统算法进行虚拟机最优分配方案挖掘的过程中,由于受到大量冗余数据的影响,造成挖掘效率降低。为此,提出一种基于蚁群算法的面向多通道Qo S需求的虚拟机分簇挖掘方法。根据虚拟机特征的相似度对虚拟机进行分簇,利用蚁群算法进行虚拟机最优分配方案的挖掘,在此过程中,充分考虑了多用户对云计算中虚拟机资源服务质量(Qo S)的要求,避免出现同类型虚拟机被分配到同一物理机上的情况,同时,对蚁群算法中信息素的更新进行了优化。实验结果表明,利用改进算法进行虚拟机最优分配方案挖掘,能够有效提高挖掘效率,并降低系统负载均衡度。A virtual machine, the optimal allocation scheme can improve the accuracy of the mining of cloud computing ser-vice quality (QoS). The optimal allocation scheme using the traditional algorithm for virtual machine in the process of min-ing, due to the influence of a large number of redundant data, which reduces the efficiency of mining. Therefore, based on ant colony algorithm is a kind of multi-channel QoS demand oriented virtual machine clustering mining method. According to the characteristics of the virtual machine similarity clustering on the virtual machine, use of ant colony algorithm for the optimal allocation of digging, the virtual machine in the process, fully consider the multi-user virtual machine in the cloud computing resources, the quality of service (QoS) requirements, to avoid the same type of virtual machine was assigned to the same physical machine, at the same time, the update of pheromone in ant colony algorithm is optimized. Experimental results show that the improved algorithm is the optimal allocation scheme for virtual machine mining, can effectively im-prove the efficiency of mining, and reduce the system load balancing degree.

关 键 词：多通道 QOS 虚拟机分簇 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]