S-Bridge:性能非对称多核处理器下负载均衡代理机制
作者:
作者简介:

赵姗(1982-),女,高级工程师,CCF学生会员,主要研究领域为操作系统,软硬件深度融合;翟健(1981-),男,博士,高级工程师,CCF专业会员,主要研究领域为软件工程,知识工程,操作系统;郝春亮(1986-),男,博士,助理研究员,主要研究领域为操作系统,集群计算,人工智能;李明树(1966-),男,博士,研究员,博士生导师,CCF会士,主要研究领域为操作系统深度设计(包括安全操作系统,数据操作系统等),可信软件过程,基础软硬件核心技术与应用.

通讯作者:

郝春亮,E-mail:chunliang@iscas.ac.cn

基金项目:

国家自然科学基金(61305054);中国科学院战略性先导科技专项基金(XDA-Y01-01)


S-Bridge: CPU Load Balancing Agent for Performance Asymmetric Multicore Processors
Author:
Fund Project:

National Natural Science Foundation of China (61305054); Strategic Priority Research Program of Chinese Academy of Sciences (XDA-Y01-01)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [40]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    近年来,在移动计算环境中,异构多核处理器已经逐渐成为主流.与传统同构的处理器设计相比,此类异构多核处理器以更低的功耗成本满足设备的计算需求.但是异构环境下CPU核之间的微架构差异,也为操作系统中的一些基本方法提出了新的挑战.面向性能非对称异构多核环境下调度的负载均衡问题,从系统层面提出了一种负载均衡机制S-Bridge,可以减少处理器微架构差异以及任务执行需求差异对传统负载均衡带来的影响.S-Bridge的主要贡献是从系统层提供了通用的、适配异构性的负载均衡相关接口,使任意调度器都能方便地与异构多核处理器系统进行适配.基于CFS和HMP调度器在ARM平台上进行实验,同时在X86平台上进行S-Bridge通用性的验证,结果表明:S-Bridge可以支持不同真实平台和内核版本的快速实现,平均性能提升超过15%,部分情况下可达65%.

    Abstract:

    In recent years, heterogeneous multi-core processors have gradually become the mainstream in the mobile computing environment. Compared with the traditional processor design, they can meet the computing needs of devices at a lower power cost. Microarchitecture differences between the CPU cores also pose new challenges for some basic methods in the operating systems. In this study, in order to resolve the load balancing problem of heterogeneous scheduling, a new load balancing mechanism called S-Bridge is proposed, which reduces the influence of the processor microarchitecture and the task requirement diversity. The main contribution of S-Bridge is to provide a universal, heterogeneity-aware load balancing interface, so that any scheduler can easily adapt to the heterogeneous multi-core processor systems. The experiments based on CFS and HMP on the X86 and ARM platforms show that S-Bridge can be implemented on different platforms with different kernel versions. The average performance increases by more than 15%, and in some best cases 65% is achieved.

    参考文献
    [1] Mittal S. A survey of techniques for architecting and managing asymmetric multicore processors. ACM Computing Surveys, 2016, 48(3):1-38.
    [2] Kumar R, Farkas KI, Jouppi NP, et al. Single-ISA heterogeneous multi-core architectures:The potential for processor power reduction. In:Proc. of the 36th Annual IEEE/ACM Int'l Symp. on Microarchitecture. 2003.81-92.
    [3] Kumar R, Tullsen DM, Ranganathan P, et al. Single-ISA heterogeneous multi-core architectures for multithreaded workload performance. In:Proc. of the ACM SIGARCH Computer Architecture News. 2004.64.
    [4] Greenhalgh P. Big.LITTLE processing with ARM CortexTM-A15& Cortex-A7. In:Proc. of the ARM. 2011.1-8.
    [5] Shiu E, Prakash S. System challenges and hardware requirements for future consumer devices:From wearable to ChromeBooks and devices in-between. In:Proc. of the IEEE 2015 Symp. on VISI Technology. 2015.1-5.
    [6] Lü F, Cui HM, Huo W, et al. Survey of scheduling policies for co-run degradation. Journal of Computer Research and Development, 2014,51(1):17-30(in Chinese with English abstract).
    [7] Koufaty D, Reddy D, Hahn S. Bias scheduling in heterogeneous multi-core architectures. In:Proc. of the 5th European Conf. on Computer Systems. 2010.125-138.
    [8] Srinivasan S, Kurella N, Koren I, et al. Exploring heterogeneity within a core for improved power efficiency. IEEE Trans. on Parallel and Distributed Systems, 2016,27(4):1057-1069.
    [9] McKenny P. A big.LITTLE scheduler update. 2012. https://lwn.net/Articles/501501/
    [10] Tseng PH, Hsiu PC, Pan CC, et al. User-Centric energy-efficient scheduling on multi-core mobile devices. In:Proc. of the 51st Annual Design Automation Conf. (DAC 2014). 2014.1-6.
    [11] Van Craeynest K, Jaleel A, Eeckhout L, et al. Scheduling heterogeneous multi-cores through performance impact estimation (PIE). In:Proc. of the ACM SIGARCH Computer Architecture News. 2012.213-224.
    [12] Chen J, Nair AA, John LK. Predictive heterogeneity-aware application scheduling for chip multiprocessors. IEEE Trans. on Computers, 2014,63(2):435-447.
    [13] Nie P, Duan Z. Efficient and scalable scheduling for performance heterogeneous multicore systems. Journal of Parallel and Distributed Computing, 2012,72(3):353-361.
    [14] Saez JC, Prieto M, Fedorova A, et al. A Comprehensive Scheduler for Asymmetric Multicore Systems. New York:Assoc Computing Machinery, 2010.139-152.
    [15] Wang T, An H, Sun T, et al. Fair scheduling on dynamic heterogeneous chipmultiprocessor. Ruan Jian Xue Bao/Journal of Software, 2014,25(Suppl.(2)):80-89(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/14026.htm
    [16] Rehman M, Asfand-E-Yar M. Scheduling on heterogeneous multi-core processors using stable matching algorithm. Int'l Journal of Advanced Computer Science and Applications, 2016,7(6).
    [17] Liu GS, Park J, Marculescu D. Dynamic thread mapping for high-performance, power-efficient heterogeneous many-core systems. In:Proc. of the IEEE 31st Int'l Conf. on Computer Design (ICCD). 2013.54-61.
    [18] Shelepov D, Saez Alcaide JC, Jeffery S, et al. HASS:A scheduler for heterogeneous multicore systems. ACM SIGOPS Operating Systems Review, 2009,43(2):66-75.
    [19] Becchi M, Crowley P. Dynamic thread assignment on heterogeneous multiprocessor architectures. In:Proc. of the 3rd Conf. on Computing Frotiers. 2006.29.
    [20] Li T, Baumberger D, Koufaty DA, et al. Efficient operating system scheduling for performance-asymmetric multi-core architectures. In:Proc. of the 2007 ACM/IEEE Conf. on Supercomputing. 2007.53.
    [21] Balakrishnan S, Rajwar R, Upton M, et al. The impact of performance asymmetry in emerging multicore architectures. In:Proc. of the 32nd Int'l Symp. on Computer Architecture. 2015.506-517.
    [22] Saez JC, Pousa A, Giusti AED, et al. On the interplay between throughput, fairness and energy efficiency on asymmetric multicore processors. Computer Journal, 2018,61(1):74-94.
    [23] Kim YG, Kim M, Chung SW. Enhancing energy efficiency of multimedia applications in heterogeneous mobile multicore processors. IEEE Trans. on Computers, 2017,P(99):1.
    [24] Tavana MK, Hajkazemi MH, Pathak D, et al. ElasticCore:A dynamic heterogeneous platform with joint core and voltage/frequency scaling. IEEE Trans. on Very Large Scale Integration Systems, 2018:1-13.
    [25] Kim M, Kim K, Geraci JR, et al. Utilization-Aware load balancing for the energy efficient operation of the big.LITTLE processor. In:Proc. of the Design, Automation & Test in Europe Conf. & Exihibition. 2014.
    [26] Petrucci V, Loques O, Mossé D, et al. Energy-Efficient thread assignment optimization for heterogeneous multicore systems. ACM Trans. on Embedded Computing Systems, 2015,14(1):1-26.
    [27] Saez JC, Pousa A, Castro F, et al. ACFS:A completely fair scheduler for asymmetric single-ISA multicore systems. In:Proc. of the 30th Annual Acm Symp. on Applied Computing, Vols. I and Ii. 2015.2027-2032.
    [28] Corbet J. Per-Entity load tracking. 2013. https://lwn.net/Articles/531853/
    [29] Intel I. Intel 64 and IA-32 architectures software developer's manual. Volume 3A:System Programming Guide, Part, 2013,1:64
    [30] Eyerman S, Eeckhout L, Karkhanis T, et al. A top-down approach to architecting CPI component performance counters. IEEE Micro, 2007,27(1):84-93.
    [31] Henning, John L. SPEC CPU2006 benchmark descriptions. ACM SIGARCH Computer Architecture News, 2006,34(4).
    [32] Guthaus M, Ringenberg J, Ernst D, et al. MiBench:A free, commercially representative embedded benchmark suite. In:Proc. of the Workload Characterization (WWC 2001). 2001.3-14.
    [33] Brodowski D, Wysocki RJ, Kumar V. CPU frequency and voltage scaling code in the Linux(TM) kernel. https://www.kernel.org/doc/Documentation/cpu-freq/governors.txt
    [34] Knauerhase R, Brett P, Hohlt B, et al. Using OS observations to improve performance in multicore systems. IEEE Micro, 2008.
    [35] Kunio U, Fumio A, Hironori K, et al. Heterogeneous Multicore Processor Technologiesfor Embedded Systems. New York:Springer Science+Business Media, 2012.
    [36] Venkat A, Tullsen DM. Harnessing ISA diversity:Design of a heterogeneous-ISA chip multiprocessor. ACM SIGARCH Computer Architecture News, 2014.121-132.
    [37] Lee W, Sunwoo D, Emmons CD, et al. Exploring heterogeneous-ISA core architectures for high-performance and energy-efficient mobile SoCs. In:Proc. of the GLSVLSI. 2017.
    附中文参考文献:
    [6] 吕方,崔慧敏,霍玮,冯晓兵.面向并发性能下降的调度策略的综述.计算机研究与发展,2014,51(1):17-30.
    [15] 王涛,安虹,孙涛,高晓川,张海博,程亦超,彭毅.面向动态异构多核处理器的公平调度算法.软件学报,2014,25(S2):80-89. http://www.jos.org.cn/1000-9825/14026.htm
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

赵姗,郝春亮,翟健,李明树. S-Bridge:性能非对称多核处理器下负载均衡代理机制.软件学报,2020,31(9):2965-2979

复制
分享
文章指标
  • 点击次数:1812
  • 下载次数: 4324
  • HTML阅读次数: 1492
  • 引用次数: 0
历史
  • 收稿日期:2018-08-04
  • 最后修改日期:2018-11-15
  • 在线发布日期: 2020-04-21
  • 出版日期: 2020-09-06
文章二维码
您是第19778409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