(3.238.250.105) 您好!臺灣時間:2021/04/20 06:17
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:杜偉瑜
研究生(外文):Wei-yu Tu
論文名稱:多工系統中個別工作耗電量分析工具
論文名稱(外文):A Task-Based Energy Consumption Analysis Tool in Multi-Tasking Systems
指導教授:羅習五
指導教授(外文):Shi-wu Lo
學位類別:碩士
校院名稱:國立中正大學
系所名稱:資訊工程所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2006
畢業學年度:94
語文別:英文
論文頁數:40
中文關鍵詞:嵌入式系統耗電量硬體量測
外文關鍵詞:embedded systemphysical measurementpower consumption
相關次數:
  • 被引用被引用:1
  • 點閱點閱:205
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:16
  • 收藏至我的研究室書目清單書目收藏:0
目前要評估一個嵌入式系統執行程式所消耗的電量,通常是使用模擬或是硬體量測的方式。模擬器可以提供很詳細的程式執行行為和耗電量,但是它的速度慢、準確度低,較適合用在無法取得系統硬體的情況下估算系統的耗電量趨勢。硬體量測是一種快速且準確的量測方式,但是量測的結果和被量測的程式之間必須要有準確的對應。另外,目前的成果較少把作業系統的行為,包括行程的切換、中斷、系統函式等因素考慮進去,而且目前已有的成果取樣率並不高,直接影響到量測結果的準確度。在本論文中,我們將硬體量測的方式和系統的執行資訊結合起來。當量測一個程式區段時,我們使用一個額外負擔很小的方式來記錄作業系統中每一個行程執行的行為,同時我們解決高取樣率的問題,用以量測嵌入式系統的耗電量來改善現有的硬體量測方式。另外,我們實作了一個分析程式負責分析作業系統和硬體量測所產生的記錄檔,以提供該量測區段的排程結果和每一個行程的執行資訊與準確的耗電量。我們相信這個工具能夠幫助低弁茠漪膍s者更有效率地去分析與評估他們的研究成果。
Simulation and physical measurement are methods often used to evaluate the energy consumption of an embedded system. Simulators can provide the detailed program behavior in both energy and performance, but they are very slow and inaccurate compared with physical measurement and only suit to estimate the energy when no concrete architecture is available. Physical measurement is fast and objective because the energy is calculated based on the current flowing through the system. However, physical measurement lacks a semantic connection between the measurement result and the program evaluated, and it seldom considers the execution activities from operating system including context switch, interrupt, system process, etc. Besides, the sample rate of recent researches is not high enough, such that it will affect the measurement accuracy. In this paper, we bring together the physical measurement and execution semantics. When profiling a program segment, we record the execution activities of every process in operating system with low overhead and measure the energy of embedded system in high sample rate to achieve improved physical measurement. We also implement an analysis tool to analyze the data from kernel and physical measurement to provide an overall system information, which includes scheduling result with power information and a profile of energy usage by processes. We believe the tool can help researchers in power consumption optimization to evaluate their work more efficiently and accurately.
1 Introduction
2 Related Work and Background
2.1 Simulation-Based Profiling Tools
2.2 PhysicalMeasurement-Based Profiling Tools
3 Energy Consumption Analysis Tool
3.1 Design Goal
3.2 The Profiling Procedure
3.2.1 The Pre-Process Stage
3.2.2 TheMeasurement Stage
3.2.3 The Analysis Stage
3.3 Implementations
3.3.1 Hardware Platform
3.3.2 Sample Rate and Accuracy
3.3.3 Measured Interval
3.3.4 Record Process Execution Data
3.3.5 Acquisition Program and Analysis Program
4 Measurement Example
5 Conclusion and Future Work
Bibliography
[1] David Brooks, Vivek Tiwari, and Margaret Martonosi. Wattch: a framework
for architectural-level power analysis and optimizations. In proceedings of the 27th Annual International Symposium on Computer Architecture, page 83, June 2000.
[2] Chen-Ting Chung and Tei-Wei Kuo. Energy profiling for mobile computing: a quantitative study of a web browser. Master’s thesis, Department of Computer Science and Information Engineering National Taiwan University, 2003.
[3] Robert P. Dick, Ganesh Lakshminarayana, Anand Raghunathan, and Niraj K. Jha. Power analysis of embedded operating systems. In Proceedings of the 37th conference on Design Automation, pages 312–315, June 2000.
[4] J. Flinn and M. Satyanarayanan. Powerscope: A tool for profiling the energy usage of mobile applications. In Proceedings of IEEE Workshop Mobile Comput. Syst. Applicat., pages 2–10, Feb 1999.
[5] Microtime Computer homepage. http://www.microtime.com.tw/.
[6] National Instrument homepage. http://www.ni.com/.
[7] Tom’s Hardware homepage. http://www.tomshardware.com/.
[8] Chuling Hu, Danial A. Jimenez, and Ulrich Kremer. Toward an evaluation infrastructure for power and energy optimizations. In proceedings of the 19th IEEE Internationl Parallel and Distributed Processing Symposium, 2005.
[9] Nam Sung Kim, Todd Austin, Trevor Mudge, and Dirk Grunwald. Challenges for Architectural Level Power Modeling. Power aware computing. Kluwer Academic Publishers, 2002.
[10] Erik Mouw. Linux kernel procfs guide. June 2001.
[11] T. Sherwood, E. Perelman, and B. Calder. Basic block distribution analysis to find periodic behavior and simulation points in applications. In International Conference on Parallel Architectures and Compilation Techniques,
pages 3–14, September 2001.
[12] Dongkun Shin, Hojun Shim, Yongsoo Joo, Han-Saem Yun, Jihong Kim, and Naehyuck Chang. Energy-monitoring tool for low-power embedded programs. IEEE Design and Test of Computers, 19:7–17, July-Aug 2002.
[13] Sim-Panalyzer. http://www.eecs.umich.edu/ panalyzer.
[14] T. K. Tan, A. Raghunathan, and N. K. Jha. A simulation framework for energy-consumption analysis of os-driven embedded applications. IEEE Transactions on computer-aided design of integrated circuits and systems, 22(9):1284–1294, Sept 2003.
[15] Bokyung Wang and Suresh Singh. Computational energy cost of tcp. In INFOCOM 2004. Twenty-third AnnualJoint Conference of the IEEE Computer and Communications Societies, volume 2, pages 785–795, March 2004.
[16] Qiang Wu, V.J. Reddi, Youfeng Wu, Jin Lee, Dan Connors, David Brooks, Margaret Martonosi, and DouglasW. Clark. A dynamic compilation framework for controlling microprocessor energy and performance. In Microarchitecture, 2005. MICRO-38. Proceedings. 38th Annual IEEE/ACM International Symposium, Nov 2005.
[17] W. Ye, N. Vijaykrishnan, M. Kandemir, and M. J. Irwin. The design and use of simplepower: a cycle-accurate energy estimation tool. In Proceedings of the 37th conference on Design automation, pages 340–345, June 2000.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊
 
系統版面圖檔 系統版面圖檔