›› 2018,Vol. 33 ›› Issue (1): 131-144.doi: 10.1007/s11390-018-1811-x

所属专题： Computer Architecture and Systems

• Special Section on Selected Paper from NPC 2011 • 上一篇    下一篇

基于等效资源配置的数据中心效能优化方法研究

Fa-Qiang Sun1,2,3, Student Member, CCF, IEEE, Gui-Hai Yan1,3,*, Member, CCF, ACM, IEEE, Xin He1,3, Student Member, CCF, IEEE, Hua-Wei Li1,3,*, Distinguished Member, CCF, Senior Member, IEEE, Member, ACM, Yin-He Han1,3, Distinguished Member, CCF, Senior Member, IEEE, Member, ACM   

  1. 1 State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences Beijing 100190, China;
    2 National Computer Network Emergency Response Technical Team of China, Beijing 100029, China;
    3 University of Chinese Academy of Sciences, Beijing 100049, China
  • 收稿日期:2016-08-09 修回日期:2017-04-20 出版日期:2018-01-05 发布日期:2018-01-05
  • 通讯作者: Gui-Hai Yan, Hua-Wei Li E-mail:yan@ict.ac.cn;lihuawei@ict.ac.cn
  • 作者简介:Fa-Qiang Sun received his B.S. degree from Heilongjiang University, Harbin, in 2010, and his Ph.D. degree in computer science from Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), Beijing, in 2017. His scientific interests include computer architecture, green computing, parallel programming, and cloud computing. He is a student member of CCF and IEEE.
  • 基金资助:

    This work is supported by the National Natural Science Foundation of China under Grant Nos. 61572470, 61532017, 61522406, 61432017, 61376043, 61504153, and 61521092, and in part by Youth Innovation Promotion Association, Chinese Academy of Sciences (CAS), under Grant No. Y404441000.

CPicker: Leveraging Performance-Equivalent Configurations to Improve Data Center Energy Efficiency

Fa-Qiang Sun1,2,3, Student Member, CCF, IEEE, Gui-Hai Yan1,3,*, Member, CCF, ACM, IEEE, Xin He1,3, Student Member, CCF, IEEE, Hua-Wei Li1,3,*, Distinguished Member, CCF, Senior Member, IEEE, Member, ACM, Yin-He Han1,3, Distinguished Member, CCF, Senior Member, IEEE, Member, ACM   

  1. 1 State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences Beijing 100190, China;
    2 National Computer Network Emergency Response Technical Team of China, Beijing 100029, China;
    3 University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2016-08-09 Revised:2017-04-20 Online:2018-01-05 Published:2018-01-05
  • Contact: Gui-Hai Yan, Hua-Wei Li E-mail:yan@ict.ac.cn;lihuawei@ict.ac.cn
  • About author:Fa-Qiang Sun received his B.S. degree from Heilongjiang University, Harbin, in 2010, and his Ph.D. degree in computer science from Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), Beijing, in 2017. His scientific interests include computer architecture, green computing, parallel programming, and cloud computing. He is a student member of CCF and IEEE.
  • Supported by:

    This work is supported by the National Natural Science Foundation of China under Grant Nos. 61572470, 61532017, 61522406, 61432017, 61376043, 61504153, and 61521092, and in part by Youth Innovation Promotion Association, Chinese Academy of Sciences (CAS), under Grant No. Y404441000.

PDF (PC)

626

服务器的低效能是导致数据中心效能低的主要原因之一.研究发现应用程序在不同的资源配置下表现出相似的性能但迥异的能耗差异.我们把这种资源配置称为等效资源配置（PERC）.基于这种发现，一种提高服务器效能的直观方法是：在做资源分配时，资源分配器为每一个应用程序选取能耗最小的资源配置.然而受限于服务器的资源容量，资源分配器不能够为每一应用选取最优的资源配置.本文我们提出一种基于遗传算法的启发式方法CPicker.CPicker使用贪心策略生成一个优良初始种群，从而加速收敛速度.实验表明，与已有的贪心算法相比，CPicker获得了17%的效能提高，而且与穷举算法相比，仅有4%的效能损失.

Abstract: The poor energy proportionality of server is seen as the principal source for low energy efficiency of modern data centers. We find that different resource configurations of an application lead to similar performance, but have distinct energy consumption. We call this phenomenon as "performance-equivalent resource configurations (PERC)", and its performance range is called equivalent region (ER). Based on PERC, one basic idea for improving energy efficiency is to select the most efficient configuration from PERC for each application. However, it cannot support every application to obtain optimal solution when thousands of applications are run simultaneously on resource-bounded servers. Here we propose a heuristic scheme, CPicker, based on genetic programming to improve energy efficiency of servers. To speed up convergence, CPicker initializes a high quality population by first choosing configurations from regions that have high energy variation. Experiments show that CPicker obtains above 17% energy efficiency improvement compared with the greedy approach, and less than 4% efficiency loss compared with the oracle case.

[1] Kirk D B, Strosnider J K. Smart (strategic memory allocation for real-time) cache design using the MIPS R3000. In Proc. the 11th Real-Time Systems Symp., December 1990, pp.322-330.

[2] Ma J Y, Sui X F, Sun N H, Li Y P, Yu Z H, Huang B W, Xu T N, Yao Z C, Chen Y, Wang H B, Zhang L X, Bao Y G. Supporting differentiated services in computers via programmable architecture for resourcing-on-demand (PARD). In Proc. the 20th Int. Conf. Architectural Support for Programming Languages and Operating Systems, March 2015, pp.131-143.

[3] Grandl R, Ananthanarayanan G, Kandula S, Rao S, Akella A. Multi-resource packing for cluster schedulers. In Proc. the 2014 ACM Conf. SIGCOMM, August 2014, pp.455-466.

[4] Zaharia M, Chowdhury M, Franklin M J, Shenker S, Stoica I. Spark:Cluster computing with working sets. In Proc. the 2nd USENIX Conf. Hot Topics in Cloud Computing, June 2010.

[5] Zaharia M, Chowdhury M, Das T, Dave A, Ma J, McCauley M, Franklin M J, Shenker S, Stoica I. Resilient distributed datasets:A fault-tolerant abstraction for in-memory cluster computing. In Proc. the 9th USENIX Conf. Networked Systems Design and Implementation, April 2012.

[6] Lee S, Panigrahy R, Prabhakaran V, Ramasubramanian V, Talwar K, Uyeda L, Wieder U. Validating heuristics for virtual machines consolidation. https://www.microsoft.com/en-us/research/wp-content/uploads/2011/01/virtualization.pdf, July 2017.

[7] Fréville A. The multidimensional 0-1 knapsack problem:An overview. European Journal of Operational Research, 2004, 155(1):1-21.

[8] Isci C, Buyuktosunoglu A, Cher C Y, Bose P, Martonosi M. An analysis of efficient multi-core global power management policies:Maximizing performance for a given power budget. In Proc. the 39th Annual IEEE/ACM Int. Symp. Microarchitecture, December 2006, pp.347-358.

[9] Nathuji R, Schwan K. VirtualPower:Coordinated power management in virtualized enterprise systems. In Proc. the 21st ACM SIGOPS Symp. Operating Systems Principles, October 2007, pp.265-278.

[10] Isci C, McIntosh S, Kephart J, Das R, Hanson J, Piper S, Wolford R, Brey T, Kantner R, Ng A, Norris J, Traore A, Frissora M. Agile, efficient virtualization power management with low-latency server power states. In Proc. the 40th Annual Int. Symp. Computer Architecture, June 2013, pp.96-107.

[11] Lo D, Cheng L Q, Govindaraju R, Barroso L A, Kozyrakis C. Towards energy proportionality for large-scale latencycritical workloads. In Proc. the 41st Annual Int. Symp. Computer Architecuture, June 2014, pp.301-312.

[12] Meisner D, Gold B T, Wenisch T F. PowerNap:Eliminating server idle power. In Proc. the 14th Int. Conf. Architectural Support for Programming Languages and Operating Systems, March 2009, pp.205-216.

[13] Meisner D, Wenisch T F. DreamWeaver:Architectural support for deep sleep. In Proc. the 17th Int. Conf. Architectural Support for Programming Languages and Operating Systems, March 2012, pp.313-324.

[14] Liu Y P, Draper S C, Kim N S. SleepScale:Runtime joint speed scaling and sleep states management for power efficient data centers. In Proc. the 41st Int. Symp. Computer Architecture (ISCA), June 2014, pp.313-324.

[15] Liu F M, Zhou Z, Jin H, Li B, Li B C, Jiang H B. On arbitrating the power-performance tradeoff in SaaS clouds. IEEE Trans. Parallel and Distributed Systems, 2014, 25(10):2648-2658.

[16] Delimitrou C, Kozyrakis C. Quasar:Resource-efficient and QoS-aware cluster management. In Proc. the 19th Int. Conf. Architectural Support for Programming Languages and Operating Systems, March 2014, pp.127-144.

[17] Delimitrou C, Kozyrakis C. Paragon:QoS-aware scheduling for heterogeneous datacenters. In Proc. the 18th Int. Conf. Architectural Support for Programming Languages and Operating Systems, March 2013, pp.77-88.

[18] Lo D, Cheng L Q, Govindaraju R, Ranganathan P, Kozyrakis C. Heracles:Improving resource efficiency at scale. In Proc. the 42nd Annual Int. Symp. Computer Architecture, June 2015, pp.450-462.

[19] Yang H L, Breslow A, Mars J, Tang L J. Bubble-flux:Precise online QoS management for increased utilization in warehouse scale computers. In Proc. the 40th Annual Int. Symp. Computer Architecture, June 2013, pp.607-618.

[20] Beloglazov A, Buyya R. Adaptive threshold-based approach for energy-efficient consolidation of virtual machines in cloud data centers. In Proc. the 8th Int. Workshop on Middleware for Grids, Clouds and e-Science, December 2010.

[21] Salimian L, Safi F. Survey of energy efficient data centers in cloud computing. In Proc. the 6th Int. Conf. Utility and Cloud Computing, December 2013, pp.369-374.

[22] Xu F, Liu F M, Liu L H, Jin H, Li B, Li B C. iAware:Making live migration of virtual machines interference-aware in the cloud. IEEE Trans. Computers, 2014, 63(12):3012-3025.

[23] Deng W, Liu F M, Jin H, Liao X F, Liu H K. Reliabilityaware server consolidation for balancing energy-lifetime tradeoff in virtualized cloud datacenters. International Journal of Communication Systems, 2014, 27(4):623-642.

[24] Baset S A, Wang L, Tang C Q. Towards an understanding of oversubscription in cloud. In Proc. the 2nd USENIX Conf. Hot Topics in Management of Internet, Cloud, and Enterprise Networks and Services, April 2012.

[25] Householder R, Arnold S, Green R. On cloud-based oversubscription. International Journal of Engineering Trends and Technology (IJETT), 2014, 8(8):425-431.

[26] Wang L, Zhan J F, Luo C J, Zhu Y Q, Yang Q, He Y Q, Gao W L, Jia Z, Shi Y J, Zhang S J, Zheng C, Lu G, Zhan K, Li X N, Qiu B Z. BigDataBench:A big data benchmark suite from Internet services. In Proc. the 20th Int. Symp. High Performance Computer Architecture (HPCA), February 2014, pp.488-499.

[27] Yan G H, Ma J, Han Y H, Li X W. EcoUp:Towards economical datacenter upgrading. IEEE Trans. Parallel and Distributed Systems, 2016, 27(7):1968-1981.

[28] Chen T S, Guo Q, Temam O, Wu Y, Bao Y G, Xu Z W, Chen Y J. Statistical performance comparisons of computers. IEEE Trans. Computers, 2015, 64(5):1442-1455.

[29] Yan G H, Sun F Q, Li H W, Li X W. CoreRank:Redeeming "Sick Silicon" by dynamically quantifying core-level healthy condition. IEEE Trans. Computers, 2016, 65(3):716-729.

[30] Winter J A, Albonesi D H, Shoemaker C A. Scalable thread scheduling and global power management for heterogeneous many-core architectures. In Proc. the 19th Int. Conf. Parallel Architectures and Compilation Techniques, September 2010, pp.29-40.

[31] Nia M B, Alipouri Y. Speeding up the genetic algorithm convergence using sequential mutation and circular gene methods. In Proc. the 9th Int. Conf. Intelligent Systems Design and Applications, December 2009, pp.31-36.

[32] Kansal A, Zhao F, Liu J, Kothari N, Bhattacharya A A. Virtual machine power metering and provisioning. In Proc. the 1st ACM Symp. Cloud Computing, June 2010, pp.39-50.
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 刘明业; 洪恩宇;. Some Covering Problems and Their Solutions in Automatic Logic Synthesis Systems[J]. , 1986, 1(2): 83 -92 .
[2] 陈世华;. On the Structure of (Weak) Inverses of an (Weakly) Invertible Finite Automaton[J]. , 1986, 1(3): 92 -100 .
[3] 高庆狮; 张祥; 杨树范; 陈树清;. Vector Computer 757[J]. , 1986, 1(3): 1 -14 .
[4] 陈肇雄; 高庆狮;. A Substitution Based Model for the Implementation of PROLOG——The Design and Implementation of LPROLOG[J]. , 1986, 1(4): 17 -26 .
[5] 黄河燕;. A Parallel Implementation Model of HPARLOG[J]. , 1986, 1(4): 27 -38 .
[6] 闵应骅; 韩智德;. A Built-in Test Pattern Generator[J]. , 1986, 1(4): 62 -74 .
[7] 唐同诰; 招兆铿;. Stack Method in Program Semantics[J]. , 1987, 2(1): 51 -63 .
[8] 闵应骅;. Easy Test Generation PLAs[J]. , 1987, 2(1): 72 -80 .
[9] 朱鸿;. Some Mathematical Properties of the Functional Programming Language FP[J]. , 1987, 2(3): 202 -216 .
[10] 李明慧;. CAD System of Microprogrammed Digital Systems[J]. , 1987, 2(3): 226 -235 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持：support@magtech.com.cn
总访问量：