一种动态优先级排序的虚拟机I/O调度算法

doi:10.11896/j.issn.1002-137X.2017.01.003

摘要/Abstract

摘要： I/O任务调度是影响I/O密集型虚拟机性能的重要因素。现有调度方法主要是针对虚拟机整机I/O带宽的优化,较少兼顾各虚拟域与全局性能,也无法满足域间差异化服务的要求。针对现有方法的不足,提出了一种动态优先级排序的虚拟机I/O调度算法DPS。该算法基于多属性决策理论,以离差最大化方法计算I/O任务的优先级评估属性权重,对I/O任务优先级进行综合评估；通过引入任务所在虚拟域价值,体现云计算环境下虚拟域重要性差异。在Xen系统中通过实验评测DPS调度虚拟化网卡的性能,结果表明,DPS能够有效提高指定域与全局的I/O任务截止期保证率、整机I/O带宽,并能为不同虚拟域的I/O应用提供差异化服务。

关键词: 云计算,虚拟化,I/O调度,差异化服务,多属性决策

Abstract: I/O scheduling is a key factor for the performance of I/O-intensive virtual machine.The existing scheduling methods mainly focus on optimizing the global I/O bandwidth of virtual machine,and few taking the performance of each virtual domain and the entire system synchronously into account.According to the deficiencies of the existing methods,I/O scheduling algorithm DPS was proposed based on dynamic prioritization.The algorithm takes advantage of multi-attribute decision making theory to calculate attribute weights dynamically for assessing I/O task priority comprehensively,and effectively characterize virtual domain by its value.The performance of DPS is evaluated by scheduling virtualized NIC in Xen.The results show that DPS improves the deadline guarantee ratios of specified domain and the entire system,improves global I/O bandwidth,and it can provide differentiated services for each virtual domain on demand.

Key words: Cloud computing,Virtualization,I/O scheduling,Differentiated service,Multi-attribute decision making

郭松辉,龚雪容,王炜,李清宝,孙磊. 一种动态优先级排序的虚拟机I/O调度算法[J]. 计算机科学, 2017, 44(1): 13-19. https://doi.org/10.11896/j.issn.1002-137X.2017.01.003

GUO Song-hui, GONG Xue-rong, WANG Wei, LI Qing-bao and SUN Lei. I/O Scheduling Algorithm Based on Dynamic Prioritization in Virtual Machines[J]. Computer Science, 2017, 44(1): 13-19. https://doi.org/10.11896/j.issn.1002-137X.2017.01.003

参考文献

[1] XIAO P,HU Z G,LIU D B,et al.Energy-efficiency enhanced virtual machine scheduling policy for mixed workloads in cloud environments[J].Computers & Electrical Engineering,2014,40(5):1650-1665.
[2] YANG Xing,MA Zi-tang,SUN Lei.Research on Extended Ant Colony Optimization Based Virtual Machine Deployment in Infrastructure Clouds[J].Computer Science,2012,39(9):33-37.(in Chinese) 杨星,马自堂,孙磊.云环境下基于改进蚁群算法的虚拟机批量部署研究[J].计算机科学,2012,39(9):33-37 .
[3] NOORSHAMS Q,BUSCH A,RENTSCHLER A,et al.Automated Modeling of I/O Performance and Interference Effects in Virtualized Storage Systems[C]∥Proceedings of the 2014 IEEE 34th International Conference on Distributed Computing Systems Workshops (ICDCSW).Madrid,Spain:IEEE Press,2014:88-93.
[4] MYTILINIS I,TSOUMAKOS D,KANTERE V,et al.I/O Performance Modeling for Big Data Applications over Cloud Infrastructures[C]∥Proceedings of the 2015 IEEE International Conference on Cloud Engineering (IC2E).Arizona,USA:IEEE Press,2015:201-206.
[5] GUAN H B,MA R H,LI J.Workload-Aware Credit Scheduler for Improving Network I/O Performance in Virtualization Environment[J].IEEE Transactions on Cloud Computing,2014,2(2):130-142.
[6] YU C,QIN L H,ZHOU J L.A multicore periodical preemption virtual machine scheduling scheme to improve the performance of computational tasks[J].Journal of Supercomputing,2014,67(1):254-276.
[7] NOORSHAMS Q,ROSTAMI K,KOUNEV S,et al.Modeling of I/O Performance Interference in Virtualized Environments with Queueing Petri Nets[C]∥Proceedings of the 2014 IEEE 22nd International Symposium on Modelling,Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS).Paris,France:IEEE Press,2014:331-336.
[8] MUENCH D,ISFORT O,MUELLER K,et al.Hardware-BasedI/O Virtualization for Mixed Criticality Real-Time Systems Using PCIe SR-IOV[C]∥Proceedings of the 2013 IEEE 16th International Conference on Computational Science and Enginee-ring (CSE).Sydney,Australia:IEEE Press,2013:706-713.
[9] PELEG O,MORRISON A,SEREBRIN B,et al.Utilizing theIOMMU scalably[C]∥Proceedings of the 2015 USENIX Conference on Usenix Annual Technical Conference.California,USA:USENIX Association,2015:549-562.
[10] JAIN N,LAKSHMI J.PriDyn:Enabling Differentiated I/O Ser-vices in Cloud Using Dynamic Priorities[J].IEEE Transactions on Services Computing,2015,8(2):212-224.
[11] HWANG K,DONGARRA J,FOX G C.Distributed and Cloud Computing:From Parallel Processing to the Internet of Things[M].San Francisco:Margan Kaufmann,2011.
[12] ONGARO D,COX A L,RIXNER S,et al.Scheduling I/O in Virtual Machine Monitors[C]∥Proceedings of the Fourth ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’08).New York,USA:ACM Press,2008:1-10.
[13] TAN H L,HUANG L J,HE Z H,et al.DMVL:An I/O bandwidth dynamic allocation method for virtual networks[J].Journal of Network and Computer Applications,2014,39(3):104-116.
[14] BOURGUIBA M,HADDADOU K,El KORBI I,et al.Impro-ving Network I/O Virtualization for Cloud Computing[J].IEEE Transactions on Parallel and Distributed Systems,2014,25(3):673-681.
[15] WANG Yong-yan,WANG Qiang,WANG Hong-an,et al.A Real-Time Scheduling Algorithm Based on Priority Table and Its Implementation[J].Journal of Software,2004,15(3):360-370.(in Chinese) 王永炎,王强,王宏安,等.基于优先级表的实时调度算法及其实现[J].软件学报,2004,15(3):360-370.
[16] WU Z B,FANG Y F.A Consensus and Maximizing Deviationbased Approach for Multi-criteria Group Decision Making under Linguistic Setting[C]∥Proceedings of the 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE).Beijing,China:IEEE Press,2014:469-475.
[17] WANG Ying-ming.Using the Method of Maximizing Deviationsto Make Decision for Multiindicies[J].Journal of Systems Engineering and Electronics,1998(7):24-26.(in Chinese) 王应明.运用离差最大化方法进行多指标决策与排序[J].系统工程与电子技术,1998(7):24-26.
[18] LI X Y,MA H D,ZHOU F,et al.T-Broker:A Trust-Aware Service Brokering Scheme for Multiple Cloud Collaborative Services[J].IEEE Transactions on Information Forensics and Security,2015,10(7):1402-1415.
[19] BARHAM P,DRAGOVIC B,FRASER K,et al.Xen and theArt of Virtualization[C] ∥Proceedings of the 19th ACM Sympo-sium on Operating Systems Principles (SOSP ’03).New York,USA:ACM Press,2003:164-177.
[20] CHISNALL D.The Definitive Guide to the Xen Hypervisor[M].New Jersey:Prentice Hall,2007.
[21] NING F F,WENG C L,LUO Y.Virtualization I/O Optimization Based on Shared Memory[C] ∥Proceedings of the 2013 IEEE International Conference on Big Data.California,USA:IEEE Press,2013:70-77.
[22] HUANG Liang,FENG Deng-guo,LIAN Yi-feng,et al.Method of DDoS Countermeasure Selection Based on Multi-Attribute Decision Making[J].Journal of Software,2015,26(7):1742-1756.(in Chinese) 黄亮,冯登国,连一峰,等.一种基于多属性决策的DDoS防护措施遴选方法[J].软件学报,2015,26(7):1742-1756 .
[23] XIA Jia-li,CHEN Hui,YANG Bing.A Real-Time Tasks Sche-duling Algorithm Based on Dynamic Priority[J].Chinese Journal of Computers,2012,35(12):2685-2695.(in Chinese) 夏家莉,陈辉,杨兵.一种动态优先级实时任务调度算法[J].计算机学报,2012,35(12):2685-2695.
[24] WANG Y J.A fuzzy multi-criteria decision-making model based on simple additive weighting method and relative preference relation[J].Applied Soft Computing,2015,30:412-420 .

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed