作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (10): 228-230. doi: 10.3969/j.issn.1000-3428.2011.10.079

• 工程应用技术与实现 • 上一篇    下一篇

针对ARM指令集的层次分类能耗测量方法

廖海艳 1,郭 兵 1,沈 艳 2,任 磊 1,王继禾 1,伍元胜 1   

  1. (1. 四川大学计算机学院,成都 610065;2. 成都信息工程学院控制工程学院,成都 610225)
  • 出版日期:2011-05-20 发布日期:2011-05-20
  • 作者简介:廖海艳(1982-),女,硕士研究生,主研方向:嵌入式实时系统;郭 兵,教授、博士后;沈 艳,副教授、博士;任 磊,硕士研究生;王继禾、伍元胜,博士研究生
  • 基金资助:
    国家“863”计划基金资助项目(2008AA01Z105);国家自然科学基金资助项目(61073045);四川省杰出青年科技基金资助项目(2010JQ0011)

Hierarchical Classification Energy Consumption Measurement Method for ARM Instruction Set

LIAO Hai-yan 1, GUO Bing 1, SHEN Yan 2, REN Lei 1, WANG Ji-he 1, WU Yuan-sheng 1   

  1. (1. College of Computer Science, Sichuan University, Chengdu 610065, China;2. College of Control Engineering, Chengdu University of Information Technology, Chengdu 610225, China)
  • Online:2011-05-20 Published:2011-05-20

摘要: 在分析ARM指令集的寻址方式、寄存器个数、指令周期数等特征的基础上,考虑三级流水线对指令集能耗特征的影响,提出一种层次分类能耗测量方法。实验结果验证了该方法的有效性,得出指令集能耗与电流值和指令周期数的乘积呈正比,减少指令周期数能降低指令集能耗,并且3种应用程序用例在仿真平台HMSim的测量值与W90P710实际目标板的能耗值相接近,绝对误差在10%以内。

关键词: 层次分类能耗测量, 三级流水线, ARM指令集, 能耗仿真平台

Abstract: According to the analysis of addressing mode of ARM instructions, the number of registers, and the number of instructions cycles etc, this paper considers the influence of three-level pipelining to the energy and proposes a hierarchical classification energy consumption measurement method. Experimental result shows that this method can efficiently measure the power consumption of the instruction set, and concludes that the energy consumption of instruction set is nonnegative with the product of current and the instruction cycles. Reducing the cycles can reduce the energy of instruction set. The measured energy value of 3 kinds of program executed on HMSim simulator is close to the energy of executed on the target board W90P710, the absolute error is below 10%.

Key words: hierarchical classification energy consumption measurement, three-level pipelining, Advanced RISC Machine(ARM) instruction set, energy consumption simulation platform

中图分类号: