期刊文献+

温度约束多核处理器最大稳态吞吐量分析 被引量:1

Analysis of Maximum Steady-State Throughput for Temperature-Constrained Multicore Processors
下载PDF
导出
摘要 随着多核处理器功耗密度的不断增大,温度约束条件下的性能分析已经成为多核处理器早期设计优化的重要组成部分.当处理器运行不同的任务时,处理器温度具有很大的差异性,但现有研究成果并没有考虑任务差异性对处理器性能的影响.针对采用动态频率电压调节作为温度管理技术的多核处理器,为了提高在温度约束条件下稳态吞吐量的分析准确性,考虑不同任务之间的差异性,提出一种新的最大吞吐量分析方法.将任务特征引入性能分析模型,论证了当多核处理器吞吐量达到最大值时各处理器核上任务特征之间的关系,将最大稳态吞吐量分析归结为线性规划问题.仿真实验结果表明,所提方法具有较好的分析准确性,任务特征对多核处理器最大吞吐量具有非常大的影响. With the increasing power density of multicore processors,the temperature-constrained performance analysis becomes a key component for the early design optimization of multicore processors.When different tasks are running,the temperatures of processors exhibit significant variation.However,most of existing researches for the steady-state analysis are based on the assumption that all tasks have the same power dissipation and distribution,and do not consider the impact of task variation on the performance of thermal-aware multicore processors.In order to improve the analysis accuracy of the steady-state throughput of multicore processors under temperature constraint,the task variation is taken into account,and a new method of maximum throughput analysis is proposed based on the HotSpot thermal model for the multicore processors which use dynamic voltage and frequency scaling(DVFS)to dynamically manage temperature.The task characteristic is incorporated into the model of performance analysis,and the relationship among the characteristics of tasks on various cores is demonstrated when the multicore processors achieve maximum throughput.And then the analysis of maximum throughput of multicore processors under temperature constraint is transformed to the problem of linear programming.Experimental results show that the proposed method achieves better accuracy of analysis,and task characteristic has the significant impact on the maximum steady-state throughput of temperature-constrained multicore processors.
出处 《计算机研究与发展》 EI CSCD 北大核心 2015年第9期2083-2093,共11页 Journal of Computer Research and Development
基金 国家"八六三"高技术研究发展计划基金项目(2013AA01A209) 北京市科技计划项目(D141100003414002) 北京市自然科学基金项目(4142034) 北京高等学校青年英才计划项目(YETP0380) 中央高校基本科研业务费专项资金资助项目(FRF-TP-14-042A2)
关键词 多核处理器 吞吐量分析 温度约束 任务差异性 动态电压频率调整 multicore processor throughput analysis temperature constraint task variation dynamic voltage and frequency scaling(DVFS)
  • 相关文献

参考文献33

  • 1Donald J, Martonosi M. Techniques for muhicore thermal management: Classiffcation and new exploration [C] //Proc of the 33rd IEEE Int Syrup on Computer Architecture (ISCA). Piscataway, NJ: IEEE, 2006:78-88. 被引量:1
  • 2Kong J, Chung S W, Skadron K. Recent thermal management techniques for microprocessors [J]. ACM Computing Surveys, 2012, 44(3) :13. 被引量:1
  • 3吴琦,熊光泽.基于随机决策模型的动态功耗管理策略研究[J].计算机学报,2007,30(4):622-628. 被引量:10
  • 4吴小东,韩建军,王天江.一种基于VFD多核系统的硬实时任务节能调度算法[J].计算机研究与发展,2012,49(5):1018-1027. 被引量:9
  • 5张冬松,吴彤,陈芳园,金士尧.多核系统中基于G1obal EDF的在线节能实时调度算法[J].软件学报,2012,23(4):996-1009. 被引量:11
  • 6Skadron K, Stun M R, Sankaranarayanan K, et al. Temperature aware microarchitecture: Modeling and implementation [J]. ACM Trans on Architecture and Code Optimization, 2004, 1(1): 94-125. 被引量:1
  • 7Huang W, Ghosh S, Velusamy S, et al. HotSpot: A compact thermal modeling methodology for early-stage VLSI design[J]. IEEE Trans on Very Large Scale Integration (VLSI) Systems, 2006, 14(5): 501-513. 被引量:1
  • 8Li D, Tan S X D, Pacheco E H, et al. Parameterized architecture level dynamic thermal models for muiticore microprocessors [J]. ACM Trans on Design Automation of Electronic Systems, 2010, 15(2): 16. 被引量:1
  • 9Wang H, Tan S X D, Li D, et al. Composable thermal modeling and simulation for architecture-level thermal designs of multicore microprocessors [J]. ACM Trans on Design Automation oi'Electronic Systems, 2013, 18(2): 28. 被引量:1
  • 10Hanumaiah V, Rao R, Vrudhula S, et al. Throughput optimal task allocation under thermal constraints for multi- core processors [C] //Proc of the 46th Annual Design Automation Conf (DAC). New York: ACM, 2009:776-781. 被引量:1

二级参考文献54

  • 1吴琦,熊光泽.非平稳自相似业务下自适应动态功耗管理[J].软件学报,2005,16(8):1499-1505. 被引量:20
  • 2钟虓,齐勇,侯迪,苗蕾,郑晓梅.基于DVS的多核实时系统节能调度[J].电子学报,2006,34(B12):2481-2484. 被引量:7
  • 3Rele S, Pande S, Onder S, Gupta R. Optimizing static power dissipation by functional units in superscalar processors. Lecture Notes in Computer Science, 2002,2304:85-100. [doi- 10.1007/3-540-45937-5_19]. 被引量:1
  • 4Chandrakasan AP, Sheng S, Brodersen RW. Low-Power CMOS digital design. IEEE Journal of Solid-State Circuit, 1992,27(4): 473-484. [doi: 10.1109/4.126534]. 被引量:1
  • 5Kim W, Gupta MS, Wei GY, Brooks D. System level analysis of fast, per-core DVFS using on-chip switching regulators. In: Proc. of the IEEE 14th Int'l Symp. on High Performance Computer Architecture. Los Alamitos: IEEE, 2008. 123-134. [doi: 10.1109/ HPCA.2008.4658633]. 被引量:1
  • 6Pillai P, Shin KG. Real-Time dynamic voltage scaling for low-power embedded operating systems. In: Proc. of the 18th ACM Symp. on Operating Systems Principles. New York: ACM, 2001.89-102. [doi: 10.1145/502059.502044]. 被引量:1
  • 7Aydin H, Yang Q. Energy-Aware partitioning for multiprocessor real-time systems. In: Proe. of the 17th Int'l Symp. on Parallel and Distributed Processing. Los Alamitos: IEEE, 2003.113.2. [doi: 10.1109/IPDPS.2003.1213225]. 被引量:1
  • 8Chen JJ, Yang CY, Kuo TW. Slack reclamation for real-time task scheduling over dynamic voltage scaling multiprocessors. In: Proc. of the IEEE Int'l Conf. on Sensor Networks, Ubiquitous, and Trustworthy Computing. Los Alamitos: IEEE, 2006. 358-367. [doi: 10.1109/SUTC. 2006.1636201 ]. 被引量:1
  • 9Chen JJ, Hsu HR, Kuo TW. Leakage-Aware energy-efficient scheduling of real-time tasks in multiprocessor systems. In: Proc. of the IEEE Real-Time and Embedded Technology and Applications Symp. Los Alamitos: IEEE, 2006. 408-417. [doi: 10.1109/RTAS. 2006.25]. 被引量:1
  • 10Anderson JH, Baruah SK. Energy-Efficient synthesis of periodic task systems upon identical multiprocessor platforms. In: Proc. of the 24th Int'l Conf. on Distributed Computing Systems. Los Alamitos: IEEE, 2004. 428-435. [doi: 10.1109/ICDCS.2004.1281609]. 被引量:1

共引文献24

同被引文献9

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部