基于合作博弈的多虚拟机实时迁移带宽分配机制

doi:10.11959/j.issn.1000-436x.2016082

通信学报 ›› 2016, Vol. 37 ›› Issue (4): 149-158.doi: 10.11959/j.issn.1000-436x.2016082

基于合作博弈的多虚拟机实时迁移带宽分配机制

崔勇^1,²,林予松²,李润知²,王宗敏²

¹ 郑州大学信息工程学院，河南郑州450001
² 郑州大学信息网络省重点开放实验室，河南郑州450052

出版日期:2016-04-25 发布日期:2016-04-26
基金资助:
教育部博士点专项科研基金资助项目;河南省科研重点基金资助项目;河南省创新人才基金资助项目

Cooperative game based bandwidth allocation mechanism live migration of multiple virtual machines

Yong CUI^1,²,Yu-song LIN²,Run-zhi LI²,Zong-min WANG²

¹ Institute of Information Engineering, Zhengzhou University, Zhengzhou 450001, China
² Key Lab on Information Networking, Zhengzhou University, Zhengzhou 450052, China

Online:2016-04-25 Published:2016-04-26
Supported by:
The Specialized Research Fund for the Doctoral Program for Higher Education;The Key Research Project of Henan Province Department of Education;The Innovative Talent Project of Henan Province Department of Education

摘要/Abstract

摘要：

IaaS云计算平台采用虚拟机实时迁移技术进行资源动态调度和管理。在实际应用场景下，需要并行实时迁移多个虚拟机。由于实时迁移算法本身以最大利用带宽的方式进行数据传输，存在着迁移进程间竞争带宽的问题，无法保证带宽全局最优分配，影响整体迁移的性能。提出一种基于合作博弈的多虚拟机实时迁移带宽分配机制，将带宽分配问题建模为一个纳什议价，通过求解纳什议价解得到帕累托最优的带宽分配方案，并在实际的虚拟化平台上进行了实现。实验结果表明，相比标准的并行实时迁移，所提出的带宽分配机制能够公平有效地分配带宽，提高了并行实时迁移的性能。

关键词: 实时迁移, 虚拟机, 带宽分配, 纳什议价, 合作博弈

Abstract:

In IaaS cloud computing platform, live migration of virtual machines plays a significant role in resource dy-namic dispatching and management. Under many scenarios, multiple virtual machines in the same server need to be moved online concurrently.Since live migration algorithm utilizes bandwidth as more as possible to send data, there is a bandwidth competition among the concurrent migration processes, which cannot guarantee the optimal allocation for the bandwidth and finally degrades the overall performance of the parallel live migration. A cooperative game based band-width allocation mechanism in live migration of multip virtual machines was proposed, which models the bandwidth allocation problem as a Nash bargaining game and attains a desirable bandwidth sharing scheme that guarantees Pareto optimality. Experiment results show that the proposed method can assign the bandwidth fairly and efficiently and im-prove the performance of the parallel live migration.

Key words: live migration, virtual machine, bandwidth allocation, Nash bargaining, cooperative game

崔勇,林予松,李润知,王宗敏. 基于合作博弈的多虚拟机实时迁移带宽分配机制[J]. 通信学报, 2016, 37(4): 149-158.

Yong CUI,Yu-song LIN,Run-zhi LI,Zong-min WANG. Cooperative game based bandwidth allocation mechanism live migration of multiple virtual machines[J]. Journal on Communications, 2016, 37(4): 149-158.

图/表 8

图1

表1

图2

图3

图4

图5

图6

图7

参考文献 16

[1]	董健康，王洪波，李阳阳，等． IaaS 环境下改进能源效率和网络性能的虚拟机放置方法[J]．通信学报， 2014,35(1):72-81. DONG J K , WANG H B , LI Y Y , et al. Improving energy efficiency and network performance in IaaS cloud with virtual machine place-ment[J]. Journal on Communications, 2014,35(1):72-81.
[2]	KVM. Kernel based virtual machine[EB/OL]. .
[3]	CLARK C , FRASER K , HAND S , et al. Live migration of vi ual machines[C]// The Second Symposiumon Networked Systems Design and Implementation. c2005:273-286.
[4]	YE K , JIANG X , MA R , et al. VC-Migration: live migration of virtual clusters in the cloud[C]// The ACM/IEEE 13th International Confe-rence on Grid Computing. c2012:209-218.
[5]	CALLEGATI F , CERRONI W . Live migration of virtualized networks: Analytical modeling and performance evaluation[C]// IEEE Workshop on Software Defined Networks for Future Networks and Services(SDN4FNS 2013). Trento, c2013:1-6.
[6]	LIU H , JIN H , XU C Z , et al. Performance and energy modeling for live migration of virtual machines[J]. Cluster comput ng, 2013,16(2):249-264.
[7]	YA?CHE H , MAZUMDAR R R , ROSENBERG C . A game theoretic framework for bandwidth allocation and pricing in broadband net-works[J]. IEEE/ACM Transactions on Networking(TON), 2000,8(5):667-678.
[8]	FUDENBERG D , TIROLE J . Game theory[M]. MIT Press, 1991.
[9]	MUTHOO A . Bargaining theory with applications[M]. Cambridge University Press, 1999.
[10]	STEPHEN B , LIEVEN V . Convex optimization[M]. Cambridge University Press, 2004.
[11]	BERTSEKAS D . Nonlinear programming[M]. Athena Scientific, 1995.
[12]	BREITGAND D , KUTIEL G , RAZ D . Cost-aware live migration of services in the cloud[C]// The 3rd Annual Haifa Experimental Systems Conference(SYSTOR 2010). c2010.
[13]	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):1816-1825. LIU S H , SUN Y Q , LIU G Y . An adaptive bandwidth allocatio lgo-rithm for virtual machine migration based on service features[J]. Chinese Journal of Computers, 2013,36(9):1816-1825.
[14]	DESHPANDE U , WANG X , GOPALAN K . Live gang migration of virtual machines[C]// The 20th ACM International Sympos um on High Performance Distributed Computing. c2011:135-146.
[15]	LIU H , HE B . VMbuddies: coordinating live migration of multi-tier applications in cloud environments[J]. IEEE Transactio n Parallel and Distributed Systems, 2015,26(4):1192-1205.
[16]	GUO J , LIU F , ZENG D , et al. A cooperative game based allocation for sharing data center networks[C]// IEEE Internationa ference on Computer Communications(INFOCOM). c2013:2139-2147.

虚拟机		脏页率/(Mbit·s^{- 1})
虚拟机	最小值	最大值	平均值
VM₁	10	60	48
VM₂	5	23	17
VM₃	4	25	15
VM₄	1	17	6

基于合作博弈的多虚拟机实时迁移带宽分配机制

Cooperative game based bandwidth allocation mechanism live migration of multiple virtual machines

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 16

相关文章 15

Metrics

推荐阅读 0

[1]	郎磊, 王荆宁, 王一, 赵子涛. 无人机辅助通信中基于用户轨迹的无线资源和航迹优化[J]. 通信学报, 2022, 43(3): 225-232.
[2]	宋宇波, 陈琪, 宋睿, 胡爱群. 基于虚拟机字节码注入的Android应用程序隐私保护机制[J]. 通信学报, 2021, 42(6): 171-181.
[3]	何欣枫,田俊峰,刘凡鸣. 可信云平台技术综述[J]. 通信学报, 2019, 40(2): 154-163.
[4]	田俊峰, 王子龙, 何欣枫, 李珍. 基于Shamir的虚拟机放置策略[J]. 通信学报, 2019, 40(10): 90-100.
[5]	刘浩然,赵赫瑶,邓玉静,王星淇,尹荣荣. 基于非合作博弈的无线传感器网络覆盖控制算法[J]. 通信学报, 2019, 40(1): 71-78.
[6]	刘业君,刘玉莹,汉鹏超,王继东,郭磊. OFDM无源光网络中多业务分层带宽分配算法[J]. 通信学报, 2018, 39(9): 84-93.
[7]	于存谦,唐明珠,何荣希. TWDM-PON中时延约束节能动态波长带宽分配算法[J]. 通信学报, 2018, 39(9): 110-121.
[8]	田俊峰,张永超. 基于改进期望值决策法的虚拟机可信审计方法[J]. 通信学报, 2018, 39(6): 52-63.
[9]	刘维杰,王丽娜,王丹磊,尹正光,付楠. 面向云计算平台的虚拟机同驻方法[J]. 通信学报, 2018, 39(11): 116-128.
[10]	李湘,陈宁江,杨尚林,李华. 感知应用特征与网络带宽的虚拟机在线迁移优化策略[J]. 通信学报, 2017, 38(Z2): 147-155.
[11]	荣星,沈昌祥,江荣,赵勇. 基于双层非平衡散列树的云平台远程验证方案[J]. 通信学报, 2017, 38(9): 31-38.
[12]	赵硕,季新生,毛宇星,程国振,扈红超. 基于安全等级的虚拟机动态迁移方法[J]. 通信学报, 2017, 38(7): 165-174.
[13]	程子敬,赵俊楠,崔玉文,石乐义. 基于纳什议价解的地外驻留平台网络可靠性与效用代价分析[J]. 通信学报, 2017, 38(2): 10-15.
[14]	金顺福,郝闪闪,王宝帅. 融合双速率和工作休眠的虚拟机调度策略及参数优化[J]. 通信学报, 2017, 38(12): 10-20.
[15]	孟宏伟,陈钟,孟子骞. 未来网络XIA中的虚拟机跨子网迁移[J]. 通信学报, 2016, 37(3): 107-116.