一种高效的虚拟机动态内存迁移方法

doi:10.11896/j.issn.1002-137X.2016.04.022

摘要/Abstract

摘要： 内存预拷贝是虚拟机动态迁移的主流策略,但是传统内存预拷贝算法是以脏页量为工作集,这对数据修改量小但在内存中分布较广的虚拟机环境来说,总体呈现迭代次数普遍较高、迁移总时间较长和迁移数据量较大的缺点。针对这一问题,提出了基于页内脏数据的预拷贝算法,该算法用以页为单位的脏数据作为工作集,并引入了新的数据结构记录页内脏数据。相比于以脏页为工作集粒度的传统内存预拷贝算法,该算法的工作集不仅能更准确地反映虚拟机的实际迁移环境,而且粒度明显细化,使得工作集小于预先设定的阈值的几率较大,从而可以减少迁移数据冗余,降低迭代次数,缩短迁移时间,降低迁移带宽。实验表明,该算法具有较高的虚拟机动态迁移效率。

关键词: 内存迁移,页内脏偏移段,虚拟机动态迁移

Abstract: Pre-copy is the main strategy of the virtual machine live migration,but the traditional pre-copy algorithm regards dirty page as unit work set,which makes modification data small.But for a virtual machine environment which is widely distributed in memory,the overall number of iterations is generally too high,and data migration time is too long.To solve the problems,this paper proposed pre-copy algorithm based on the dirty data in page as unit work set.The algorithm uses a new data structure to record dirty data in page.Compared to the dirty pages as unit work set,the algorithm’s unit work set not only more accurately reflects the actual migration environment of a virtual machine,but also has significantly refined particle size,which makes the total work sets smaller than the preset threshold more probably,which can reduce the number of iterations,to shorten the migration time and reduce migration bandwidth.Experiments show that the algorithm has higher efficiency of virtual machine live migration.

Key words: Memory migration,Dirty offset segment within page,Live migration of virtual machines

程虹锡,谭良. 一种高效的虚拟机动态内存迁移方法[J]. 计算机科学, 2016, 43(4): 111-114. https://doi.org/10.11896/j.issn.1002-137X.2016.04.022

CHENG Hong-xi and TAN Liang. Efficient Method of Live Migration on Virtual Machine Memory[J]. Computer Science, 2016, 43(4): 111-114. https://doi.org/10.11896/j.issn.1002-137X.2016.04.022

参考文献

[1] Zhang Xin-ling,Zhang Dong,Cao Ling-ling,et al.Research of virtualization platform in Coud Computing[J].Software Guide,2013,12(11)(in Chinese) 张新玲,张东,曹玲玲,等.云计算虚拟化平台性能研究[J].软件导刊,2013,12(11)
[2] Yang Jing.Key Technologies and Optimization for dynamic migration of Virtual Machines in Cloud Computing[D].Chang sha:Central South University,2011(in Chinese) 杨旌.面向云计算的虚拟机动态迁移关键技术及优化[D].长沙:中南大学,2011
[3] Liu Peng-cheng,Chen Rong.Cloud Computing-oriented Live Migration Framework for Virtual Machine[J].Computer Enginee-ring,2010,36(5):37-39(in Chinese) 刘鹏程,陈榕.面向云计算的虚拟机动态迁移框架[J].计算机工程,2010,36(5):37-39
[4] Zhang X,Huo Z,Ma J,et al.Exploiting data deduplication to accelerate live virtual machine migration[C]∥Proceedings of the 2010 IEEE International Conference on Cluster Computing.2010:88-96
[5] Clark C,Fraser K,Hand S,et al.Live Migration of Virtual Machines[C]∥Proceedings of the 2nd Symposium on Networked Systems Design and Implementation.Boston,Massachusetts,USA,2005:273-286
[6] Clark C,Fraser K,Hand S,et al.Live migration of virtual machines[C]∥Proceedings of the 2nd Conference on Symposium on Networked Systems Design & Implementation (NSDI’05).Berkeley,USA:ACM Press,2005:273-286
[7] Hines M R,Deshpande U,Gopalan K.Post-copy live migration of virtual machines[J].ACM SIGOPS Operating Systems Review,2009,43(3):14-26
[8] Sohan R R,Moore A,Hopper A W A.Predicting the perfor-mance of virtual machine migration[C]∥Proceedings of the Modeling,Analysis & Simulation of Computer and Telecommunication Systems(MASCOTS).Florida,USA,2010:37-46
[9] Zhang Wei,Zhu Ming-fa,Gong Tao,et al.Performance degradation-aware virtual machine live migration in virtualized servers[C]∥Proceedings of the 2012 13th International Conference on Parallel and Distributed Computing,Applications and Technologies.Las Vegas,USA,2012:429-435
[10] Jin H,Deng L,Wu S,et al.Live virtual machine migration with adaptive memory compression[C]∥Proceedings of the IEEE International Conference on Cluster Computing(Cluster’09).New Orleans,Louisiana,USA,2009:1-10
[11] Svard P,Hudzia B,Tordsson J.Evaluation of delta compression techniques for efficient live migration of large virtual machines[C]∥Proceedings of the 7th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments(VEE’11).New York,2011:111-120
[12] Sun Guo-fei,Gu Jian-hua,Hu Jin-hua,et al.Improvement of live memory migration mechanism for virtual machine based on pre-copy[J].Computer Engineering,2011,37(13):36-39(in Chinese) 孙国飞,谷建华,胡金华,等.基于预拷贝的虚拟机动态内存迁移机制改进[J].计算机工程,2011,37(13):36-39
[13] Hu Liang,Zhao Jia,Xu Gao-chao,et al.HMDC:Live virtual machine migration based on hybrid memory copy and delta compression[J].Applied Mathematics & Information Sciences,2013,7(2L):639-646
[14] Liu Shi-hai,Sun Yu-qing,Liu Gu-yue.An Adaptive Bandwidth Allocation Algorithm for Virtual Machine Migration Based on Service Features[J].Chinese Journal of Computers,2013,36(9)(in Chinese) 刘诗海,孙宇清,刘古月.面向业务特征的自适应虚拟机迁移带宽分配算法[J].计算机学报,2013,36(9)
[15] Cheng Ting-wei,Zhang Pu,Zhang Zhong-qing.Memory Optimization Algorithm of Migration Based on Xen Virtual Machine[J].Computer Science,2013,40(9)(in Chinese) 陈廷伟,张璞,张忠清.基于Xen的虚拟机迁移时内存优化算法[J].计算机科学,2013,40(9)
[16] Zhang Jing-kun.Research and Application of Virtual MachineLive Migration Techniques Based on Xen[D].Shenyang:Nor-theastern University,2011(in Chinese) 张井昆.基于Xen的虚拟机动态迁移技术研究及应用[D].沈阳:东北大学,2011
[17] Chen Yang,Huai Jin-peng,Hu Chun-ming.Live Migration ofVirtual Machines Based on Hybrid Memory Copy Approach[J].Chinese Journal of Computers,2011,4(12):2278-2291(in Chinese) 陈阳,怀进鹏,胡春明.基于内存混合复制方式的虚拟机在线迁移机制[J].计算机学报,2011,4(12):2278-2291
[18] Liu Hai-kun,Jin Hai,Liao Xiao-fei,et al.Live migration of virtual machine based on full system trace and replay[C]∥Proc of the 18th ACM International Symposium on High Performance Distributed Computing.New York:ACM Press,2009:101-110

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed