期刊文献+

源程序级和算法级嵌入式软件功耗特性的分析与优化方法研究 被引量:16

Analysis and Optimization Method of Energy Consumption Characteristics in Embedded Software Based on Source-Code and Algorithm Level
下载PDF
导出
摘要 嵌入式系统的功耗优化可以在硬件和软件的多个层次进行,随着微电子技术的不断发展,各种底层先进硬件功耗优化技术的出现和应用,使得高层软件方面的功耗管理和优化技术逐步成为控制计算机系统功耗的重要手段.文中首先在完成嵌入式软件功耗度量的基础上,在硬件微结构级和电路级分析了软件功耗的产生原因和构成因素.然后,在软件源程序级和算法级两个层次上,采取相应措施改善影响嵌入式软件功耗特性的关联特征,以降低软件功耗.最后,针对"八皇后"典型算法问题进行了仿真实验,经过源程序级、算法级和源程序级与算法级混合3种优化,系统功耗最大降幅可以达到93.2%,实验结果表明软件功耗优化方法对于降低系统功耗是可行的和有效的. The energy consumption optimization of embedded systems can be undertaken in mul- tiple levels of hardware and software. Along with the rapid growth of micro-electronics technologies, various kinds of advanced low-level hardware energy consumption optimization technologies have been developed and applied, some high-level software energy consumption management and optimization technologies gradually become an important means to control the energy consumption of embedded systems. Firstly, after the energy consumption measurement of embedded soft- ware has been finished, this paper analyzes the generating reasons and constructing factors of embedded software energy consumption based on the micro-structure and circuit level of hardware. Then, in order to reduce the energy consumption of embedded software, some measures is adopted to improve the software-related factors of embedded software energy consumption characteristics based on the source-code and algorithm level of software. Finally, a C program of typical "Eight Queens" problem is optimized and compared by three methods of source-code level, algo- rithm level and mixing of source-code and algorithm level, the highest energy savings of embed-ded systems can achieve up to 93.2%, and experimental results demonstrate that the energy con- sumption optimization methods of embedded software is feasible and effective to minimize the energy consumption of embedded systems.
出处 《计算机学报》 EI CSCD 北大核心 2009年第9期1869-1875,共7页 Chinese Journal of Computers
基金 国家"八六三"高技术研究发展计划项目基金(2008AA01Z105)资助~~
关键词 嵌入式系统功耗 软件功耗 功耗优化 源程序级 算法级 embedded system energy consumption software energy consumption energy con-sumption optimization source-code level algorithm level
  • 相关文献

参考文献18

  • 1郭兵,沈艳,邵子立.绿色计算的重定义与若干探讨//中国计算机大会(CNCC2008).中国,西安,2008:44-50. 被引量:1
  • 2Tiwari Vivek, Malik Sharad, Wolfe Andrew. Power analysis of embedded software: A first step towards software power minimization. IEEE Transactions on VLSI Systems, 1994, 2(4) : 437- 445. 被引量:1
  • 3Tiwari Vivek, Malik Sharad, Wolfe Andrew. Compilation techniques for low energy: An overview//Proceedings of the IEEE Symposiumon Low Power Electronics. San Diego, CA, USA, 1994:38- 39. 被引量:1
  • 4Chung Eui-Young, Benini Luca, de Micheli Giovanni. Source code transformation based on software cost analysis//Proceeding of ISSS'01. Montreal, Quebec, Canada, 2001: 153- 158. 被引量:1
  • 5Dalai Vishal, Ravikumar C P. Software power optimization in an embedded system//Proceedings of the 14th International Conference on VLSI Design (VLSID'01). Bangalore, India, 2000:254-259. 被引量:1
  • 6Konstantakos V, Chatzigeorgiou A, Nikolaidis S, Laopoulos T. Energy consumption estimation in embedded systems. IEEE Transactions on Instrumentation and Measurement, 2008, 57(4):797-804. 被引量:1
  • 7Mukherjee T, Tang Qinghui, Ziesman C, Gupta S K S, Cayton P. Software architecture for dynamic thermal management in datacenters//Proceedings of the 2nd International Conference on Communication Systems Software and Middleware(COMSWARE'07). Bangalore, India, 2007:1-11. 被引量:1
  • 8Xian Changjiu, Cai Le, Lu Yung-Hsiang. Power measurement of software programs on computers with multiple I/O components. IEEE Transactions on Instrumentation and Measurement, 2007, 56(5):2079 -2086. 被引量:1
  • 9Aaraj Najwa, Raghunathan Anand, Ravi Srivaths, Jha Niraj K. Energy and execution time analysis of a software-based trusted platform module//Proceedings of the Design, Automation and Test in Europe Conference and Exhibition (DATE'07). Nice, France, 2007:1-6. 被引量:1
  • 10Brandolese Carlo. Source-level estimation of energy consumption and execution time of embedded software//Proceedings of the llth EUROMICRO Conference on Digital System Design Architectures, Methods and Tools (DSD'08). Parama, Italy, 2008:115-123. 被引量:1

二级参考文献58

  • 1熊志辉,李思昆,陈吉华.遗传算法与蚂蚁算法动态融合的软硬件划分[J].软件学报,2005,16(4):503-512. 被引量:87
  • 2Lu Y-H, De Micheli G. Adaptive hard disk power management on personal computers. In: Proc. of the IEEE Great Lakes Symp. on VLSI. 1999. 50-53. 被引量:1
  • 3Benini L, Bogliolo A, Paleologo GA, De Micheli G. Policy optimization for dynamic power management. IEEE Trans. on Computer-Aided Design of Integrated Circuits and Systems, 1999,18(6):813-833. 被引量:1
  • 4Qiu Q, Pedram M. Dynamic power management based on continuous-time Markov decision processes. In: Irwin MJ, ed. Proc. of the Design Automation Conf. New Orleans: ACM Press, 1999. 555-561. 被引量:1
  • 5Simunic T, Benini L, Glynn P, De Micheli G. Event-Driven power management. IEEE Trans. on Computer-Aided Design of Integrated Circuits and Systems, 2001,20(7):840-856. 被引量:1
  • 6hung E-Y, Benini L, Bogliolo A, Lu Y-H, De Micheli G. Dynamic power management for nonstationary service requests. IEEE Trans. on Computers, 2002,51(11): 1345-1360. 被引量:1
  • 7Crovella ME, Bestavros A. Self-Similarity in World Wide Web traffic: Evidence and possible causes. IEEE/ACM Trans. on Networking, 1997,5(6):835-846. 被引量:1
  • 8Taqqu MS, Willinger W, Sherman R. Proof of a fundamental result in self-similar traffic modeling. ACM Computer Communications Review, 1997. 5-23. 被引量:1
  • 9Gribble SD, Manku GS, Roselli D, Brewer EA, Gibson TJ, Miller EL. Self-Similarity in file systems. In: Gibson TJ, Gribble SD,eds. Proc. of the ACM Conf. on Measurement and Modeling of Computer Systems (SIGMETRICS). Madison: ACM Press, 1998.141-150. 被引量:1
  • 10Harchol-Balter M, Downey AB. Exploiting process lifetime distributions for dynamic load balancing. ACM Trans. on Computer Systems, 1997,15(3):253-285. 被引量:1

共引文献38

同被引文献134

引证文献16

二级引证文献45

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