论文部分内容阅读
为了实现云平台虚拟机的资源负载均衡,提出一种基于动态监测周期的动态资源管理模型.在监测模块采用基于动态监测周期的自适应资源监测算法,它根据物理服务器的资源负载状态确定动态的监测周期,提出了当结点处于轻度过载或重度过载时,物理服务器自发地执行过载均衡方法.在选择虚拟机迁移时考虑虚拟机资源的潜在增长率,提出了改进的多因素虚拟机选择算法.选择目标物理服务器时则采用降序最佳适应算法.实验结果表明,该模型能有效减少虚拟机迁移次数,降低监测机制耗能,尽可能保证了负载均衡.
In order to realize resource load balancing of cloud platform virtual machines, a dynamic resource management model based on dynamic monitoring period is proposed. The monitoring module uses a dynamic resource monitoring algorithm based on dynamic monitoring period, which determines the dynamic Monitoring period, put forward that when the node is in light overload or heavy overload, the physical server spontaneously performs the overload balancing method.Considering the potential growth rate of virtual machine resources in the choice of virtual machine migration, an improved multivariate virtual machine selection Algorithm.The best descending order adaptive algorithm is adopted when selecting the target physical server.The experimental results show that this model can effectively reduce the number of virtual machine migration and reduce the energy consumption of the monitoring mechanism to ensure the load balancing as much as possible.