A Decentralized Virtual Machine Migration Approach of Data Centers for Cloud Computing
A Decentralized Virtual Machine Migration Approach of Data Centers for Cloud Computing
复制标题
云计算数据中心分散式虚拟机迁移方法
DOI：
10.1155/2013/878542
复制
发表时间：
2013-08
影响因子：
--
通讯作者：
Jia Xuhan
中科院分区：
文献类型：
--
作者：
Wang Xiaoying;Liu Xiaojing;Fan Lihua;Jia Xuhan
随着云计算为全球范围内的大量用户提供服务，来自客户的普适应用由大型数据中心托管。在这样的平台上，采用了虚拟化技术来对底层物理资源进行多路复用。由于不同应用的传入负载差异很大，因此管理虚拟机的放置和资源分配方案对于保证服务质量是至关重要的。在本文中，我们提出了一种云计算环境下数据中心内的分布式虚拟机迁移方法。首先定义和描述了系统模型和电源模型。然后，给出了去中心化机制的关键步骤，包括负载向量的建立、负载信息的收集、虚拟机的选择和目的地的确定。为了在保证服务质量的前提下进一步节约能量消耗，提出了一种双门限分布式迁移算法。通过性能评估实验检验了该方法的效果，并对阈值等因素进行了分析和讨论。结果表明，该方法能够有效地均衡不同物理节点之间的负载，并能从整体上降低整个系统的功耗。
As cloud computing offers services to lots of users worldwide, pervasive applications from customers are hosted by large-scale data centers. Upon such platforms, virtualization technology is employed to multiplex the underlying physical resources. Since the incoming loads of different application vary significantly, it is important and critical to manage the placement and resource allocation schemes of the virtual machines (VMs) in order to guarantee the quality of services. In this paper, we propose a decentralized virtual machine migration approach inside the data centers for cloud computing environments. The system models and power models are defined and described first. Then, we present the key steps of the decentralized mechanism, including the establishment of load vectors, load information collection, VM selection, and destination determination. A two-threshold decentralized migration algorithm is implemented to further save the energy consumption as well as keeping the quality of services. By examining the effect of our approach by performance evaluation experiments, the thresholds and other factors are analyzed and discussed. The results illustrate that the proposed approach can efficiently balance the loads across different physical nodes and also can lead to less power consumption of the entire system holistically.
登录
查看更多内容
适用于多层云环境的自适应无模型资源和电源管理方法
影响因子：
3.5
作者：
Xiaoying Wang;Zhihui Du;Yinong Chen
通讯作者：
Yinong Chen
DOI：
10.1109/icacte.2010.5579340
发表时间：
2010-08
期刊：
2010 3rd International Conference on Advanced Computer Theory and Engineering(ICACTE)
影响因子：
--
作者：
Xiaoying Wang;Hui Xie;Rui Wang;Zhihui Du;Li Jin
通讯作者：
Xiaoying Wang;Hui Xie;Rui Wang;Zhihui Du;Li Jin
DOI：
10.1016/j.jss.2007.11.719
发表时间：
2008-09
期刊：
J. Syst. Softw.
影响因子：
--
作者：
Xiaoyin Wang;Zhihui Du;Yinong Chen;Sanli Li
通讯作者：
Xiaoyin Wang;Zhihui Du;Yinong Chen;Sanli Li
DOI：
--
发表时间：
2008-04
期刊：
--
影响因子：
--
作者：
Gong Chen;Wenbo He;Jie Liu;Suman Nath;Leonidas Rigas;Lin Xiao;Feng Zhao
通讯作者：
Gong Chen;Wenbo He;Jie Liu;Suman Nath;Leonidas Rigas;Lin Xiao;Feng Zhao
DOI：
10.1145/1741906.1742192
发表时间：
2010-02
期刊：
Proceedings of the International Conference and Workshop on Emerging Trends in Technology
影响因子：
--
作者：
P. Pandey;Sandeep Singh;Suraj Singh
通讯作者：
P. Pandey;Sandeep Singh;Suraj Singh