微信服务号

微信订阅号

2025年4月3日 15:38 星期四
首页 > 过刊浏览>2017年第28卷第2期 >398-410. DOI:10.13328/j.cnki.jos.005059
PDF HTML阅读 XML下载 导出引用 引用提醒
一种灵活高效的虚拟CPU调度算法
作者:
基金项目:

国家高技术研究发展计划（863）（2015AA015301，2015AA016701），武汉市应用基本研究计划（2015010101010004）


Flexible and Efficient VCPU Scheduling Algorithm
Author:
Fund Project:

National High-Tech R&D Program of China (863)(2015AA015301, 2015AA016701), Wuhan Applied Basic Research Project (2015010101010004)

  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [24]
    • |
  • 相似文献 [20]
    • | | |
  • 文章评论
    摘要:

    目前，虚拟化已经广泛应用于数据中心，但主流的虚拟CPU调度策略并没有实现对I/O性能的保障，尤其是当延时敏感型负载的虚拟机和计算敏感型负载的虚拟机竞争CPU资源时，其性能显著下降.针对上述问题，提出了一种灵活、高效的虚拟CPU调度算法FLMS（flexible I/O latency and multi-processor sensitive scheduler）.FLMS通过采用虚拟机分类、虚拟CPU绑定、多类时间片等技术降低了虚拟机的响应延时，同时基于多处理器架构重新设计了负载均衡策略，优化了虚拟CPU迁移.FLMS通用于目前主流的虚拟化方案，在软件虚拟化方式下，与最新的优化方案相比，延时降低了30%，带宽有10%的提升；在使用硬件辅助虚拟化的系统中，通过FLMS能够获得接近原生系统的I/O性能，并且保证了整个系统的公平性.

    Abstract:

    At present, virtualization technology has been widely applied in data centers. However, VCPU (virtual CPU) scheduling strategy still faces intolerable I/O delay, especially for I/O-latency sensitive VMs which suffer from significant performance degradation when competing with CPU-intensive VMs. This paper presents a flexible and efficient VCPU scheduling algorithm FLMS (flexible I/O latency and multi-processor sensitive scheduler) which utilizes VM classification, VCPU binding and flexible slicer to reduce VM response delay. The work also redesigns the load balancing strategy to ensure optimal VCPU migration. FLMS is suitable for the current mainstream virtualization solutions. It has a 30% improvement comparing with the latest software virtualization. With hardware-assisted virtualization, FLMS makes it possible for VMs to achieve near bare-metal performance and ensures the fairness of the whole system

    参考文献
    [1] Waldspurger C, Rosenblum M. I/O virtualization. Communications of the ACM, 2012,55(1):66-73.[doi:10.1145/2063176.2063194]
    [2] Smith JE, Nair R. The architecture of virtual machines. Computer, 2005,38(5):32-38.[doi:10.1109/MC.2005.173]
    [3] Rosenblum M, Garfinkel T. Virtual machine monitors:Current technology and future trends. Computer, 2005,38(5):39-47.[doi:10. 1109/MC.2005.176]
    [4] Uhlig R, Neiger G, Rodgers D, Santoni AL, Martins FCM, Anderson AV, Bennett SM, Kagi A, Leung FH, Smith L. Intel virtualization technology. Computer, 2005,38(5):48-56.[doi:10.1109/MC.2005.163]
    [5] Kutch P. PCI-SIG SR-IOV Primer:An introduction to SR-IOV technology. Application note 321211-002, Intel Corporation, 2011.
    [6] Xu C, Gamage S, Lu H, Kompella R, Xu DY. vTurbo:Accelerating virtual machine I/O processing using designated turbo-sliced core. In:Proc. of the 2013 USENIX Annual Technical Conf. (USENIX ATC 2013). Berkeley:USENIX, 2013. 243-254.
    [7] Ding XX, Xiong AP, Yang C. Optimation of Xen scheduler for multitasking. In:Proc. of the 4th IEEE Int'l Conf. on Software Engineering and Service Science (ICSESS). Piscataway:IEEE, 2013. 754-757.[doi:10.1109/ICSESS.2013.6615415]
    [8] Rao J, Wang K, Zhou X, Xu C. Optimizing virtual machine scheduling in NUMA multicore system. In:Proc. of IEEE the 19th Int'l Symp. on High Performance Computer Architecture (HPCA 2013). Piscataway:IEEE, 2013. 23-27.[doi:10.1109/HPCA.2013.6522328]
    [9] Guan HB, Ma RH, Li J. Workload-Aware credit scheduler for improving network I/O performance in virtualization environment. IEEE Trans. on Cloud Computing, 2014,2(2):130-142.[doi:10.1109/TCC.2014.2314649]
    [10] KVM homepage. http://www.linux-kvm.org/page/Main_Page
    [11] Barham P, Dragovic B, Fraser K, Hand S, Harris T, Ho A, Neugebauer R, Pratt I, Warfield A. Xen and the art of virtualization. In:Proc. of the Symp. on Operating Systems Principles (SOSP). 2003.[doi:10.1145/945445.945462]
    [12] Ackaouy E. The xen credit CPU scheduler. 2006. http://www-archive.xenproject.org/files/summit_3/sched.pdf
    [13] Shim H, Lee SM. CFS-v:I/O demand-driven VM scheduler in KVM. 2014. http://events.linuxfoundation.jp/sites/events/files/slides/kvmforum14_hshim.pdf
    [14] Jiang W, Zhou YS, Cui Y, Feng W, Chen Y, Shi YC, Wu QB. CFS optimizations to KVM threads on multi-core environment. In:Proc. of the 15th Int'l Conf. on Parallel and Distributed Systems. Piscataway:IEEE, 2009. 348-354.[doi:10.1109/ICPADS.2009. 83]
    [15] Ongaro D, Cox AL, Rixner S. Scheduling I/O in virtual machine monitors. In:Proc. of the 4th ACM SIGPLAN/SIGOPS Int'l Conf. on Virtual Execution Environments. New York:ACM, 2008. 1-10.[doi:10.1145/1346256.1346258]
    [16] Zeng LF, Wang Y, Shi W, Feng D. An improved Xen credit scheduler for I/O latency-sensitive applications on multicores. In:Proc. of the Int'l Conf. on Cloud Computing and Big Data. Piscataway:IEEE, 2013. 267-274.[doi:10.1109/CLOUDCOM-ASIA.2013. 40]
    [17] Hu Y. Long X, Zhang J, He J, Xia L. I/O scheduling model of virtual machine based on multi-core dynamical partitioning. In:Proc. of the 19th ACM Int'l Symp. on High Performance Distributed Computing. New York:ACM, 2010. 142-154.[doi:10.1145/1851476.1851494]
    [18] Xu C, Gamage S, Rao PN, Kangarlou A, Kompella RR, Xu D. vSlicer:Latency-Aware virtual machine scheduling via differentiated-frequency CPU slicing. In:Proc. of the 21st ACM Int'l Symp. on High Performance Distributed Computing. New York:ACM, 2012. 3-14.[doi:10.1145/2287076.2287080]
    [19] Cheng YX, Chen WZ, Chen X, Xu B, Zhang SY. A user-level NUMA-aware scheduler for optimizing virtual machine performance. In:Proc. of the Int'l Workshop on Advanced Parallel Processing Technologies. Berlin, Heidelberg:Springer-Verlag, 2013. 32-46.[doi:10.1007/978-3-642-45293-2_3]
    [20] https://www.devin.com/lookbusy/
    [21] http://www.iozone.org/
    [22] http://www.netperf.org/netperf/
    [23] http://home.tiscali.cz/~cz210552/webbench.html
    [24] https://launchpad.net/sysbench
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

刘珂男,童薇,冯丹,刘景宁,张炬.一种灵活高效的虚拟CPU调度算法.软件学报,2017,28(2):398-410

复制
分享
文章指标
  • 点击次数:3188
  • 下载次数: 5386
  • HTML阅读次数: 1300
  • 引用次数: 0
历史
  • 收稿日期:2015-06-24
  • 最后修改日期:2015-10-30
  • 在线发布日期: 2017-01-24
文章二维码
友情链接:中国科学院软件研究所中国计算机学会中国科学院国家自然科学基金委员会CCF数字图书馆Int'l J of Softw Info计算机系统应用
您是第19781238位访问者
版权所有：中国科学院软件研究所 京ICP备05046678号-3
地址：北京市海淀区中关村南四街4号,邮政编码：100190
电话：010-62562563 传真：010-62562533 Email：jos@iscas.ac.cn
技术支持：北京勤云科技发展有限公司

京公网安备 11040202500063号