云数据中心考虑虚拟机关联性的虚拟机放置策略

VIRTUAL MACHINE PLACEMENT STRATEGY CONSIDERING VIRTUAL MACHINE CORRELATION IN CLOUD DATA CENTER

提出云数据中心考虑虚拟机关联性的虚拟机放置策略。在物理主机状态检测和虚拟机选择阶段,采用鲁棒局部归约主机检测方法LRR(Local Regression Robust)和最小迁移时间选择方法MMT(Minimum Migration Time);在虚拟机放置阶段,采用多重相关系数来评价虚拟机之间的关联性。该策略在重新分配虚拟机的时候可以减少高关联的虚拟机被放置到同一个物理节点上的机会,尽量避免物理主机超负载问题,最终减少虚拟机迁移次数。实验结果表明:与Cloudsim中已有的虚拟机迁移办法相比,云数据中心的各类性能指标都有所改善,该实验结果对于其他企业构造节能云数据中心有很好的参考价值。

This paper proposes a virtual machine placement strategy considering virtual machine correlation in cloud data center.The Local Regression Robust(LRR)algorithm was adopted to identify critical hosts,and the Minimum Migration Time(MMT)policy was also used for selecting VMs on critical hosts to be migrated.In the virtual placement,we adopt the multiple correlation coefficient to estimate the correlation between virtual machines.VMs with low correlations are more preferred to be reallocated on the same physical host to lower the risk of overloading,and thus leading to a fewer number of migrations.The experimental results and performance analysis show that our strategy leads to a further improvement compared with the old migration strategies in Cloudsim.Our strategy is valuable for other cloud providers to build a low energy consumption cloud data center.