臺灣博碩士論文加值系統

隨著近年來行動裝置深入人們的日常生活當中，功率消耗的研究在可攜式裝置上已變成重要的系統設計議題。多數行動裝置如PDA、智慧型手機與筆記型電腦目前皆已支援動態電壓/頻率調整(Dynamic Voltage and Frequency Scaling, DVFS)，而動態電壓/頻率調整則是被公認為降低能量消耗最有效的技術之一。本論文研究即針對行動通訊裝置提出一套完整可行的硬、軟體系統架構，根據英特爾(Intel)所提的軟體框架加以修改，利用效能監控單元(Performance Monitoring Unit, PMU)監控處理器程式執行狀況，從整體系統運作觀點去動態調整電壓/頻率進而達到降低能耗的目的並延長電池的使用時間。
本研究以Creator PXA270開發板為實驗平台，執行數個效能測試程式(Benchmarks)並量測能耗，依程式特性之不同，計算出最佳能耗速度 (Critical Speed)，進而提出一即時動態調整電壓/頻率之演算法(Energy Aware - Dynamic Voltage and Frequency Scaling, EA-DVFS)然後實作於Linux 核心。研究數據顯示，使用此演算法之系統比沒有使用之系統更能有效降低20%以上的能耗。
總體而言，此研究主要探討處理器的使用率(Utilization)與能耗(Energy Consumption)間的動態關係，並進一步了解程式的特性如何影響其策略選用。

In recent years, the use of mobile devices has become highly integrated into people’s daily lives. The study of power consumption has become an important system design issue on portable devices. Many mobile systems such as PDAs, smart phones, and laptops have currently supported dynamic voltage and frequency scaling (DVFS) during the system operating, and DVFS is considered as one of most efficient techniques to decrease energy consumption. This paper demonstrates an adaptive hardware and software systems architecture on mobile devices. We modify software power management framework, which is based on Intel® XScaleTM platforms, and take advantage of monitoring the performance monitoring unit (PMU) for the usage of processor to adjust voltage and frequency dynamically in order to reduce power consumption on system, meanwhile, extend the lifetime of battery. This paper is experimented on Creator PXA270 development board, and measure relative energy consumption with several benchmarks. According to the characters of different programs, we calculate the critical speed of minimal energy consumption, and present a runtime algorithm (Energy Aware - Dynamic Voltage and Frequency Scaling, EA-DVFS), and then we implement in Linux kernel. Results of the experiments showed the energy consumption of system, which used EA-DVFS algorithm, is lower 20 percent at least than the one without the algorithm. This study mainly investigates the dynamic relationship between processor utilization and energy consumption, as well as providing a better understanding of how characters of programs related to the strategy use.

Chinese Abstract i
English Abstract ii
Acknowledgement iii
Table of Contents iv
List of Tables vi
List of Figures vii
Chapter 1 INTRODUCTION 1
1.1 Motivation 1
1.2 Contribution 2
1.3 Thesis Structure 2
Chapter 2 RELATED WORK 4
2.1 Background 4
2.2 Dynamic Voltage and Frequency Scaling 4
2.2.1 DVFS on Hard Real-Time Applications 5
2.2.2 DVFS on Soft Real-Time Applications 6
2.3 Dynamic Power Management 8
2.3.1 Timeout Policy 10
2.3.2 Predictive Technique 10
2.3.3 Stochastic Process 10
Chapter 3 ENERGY AWARE SYSTEM DESIGN 11
3.1 Terminology and System Model 11
3.2 Energy Aware DVFS (EA-DVFS) 17
Chapter 4 SYSTEM ARCHITECTURE 20
4.1 Hardware Power Management Mechanism 20
4.1.1 Functional Overview 20
4.1.2 Power-Consumption Specifications 22
4.1.3 Performance Monitioring Unit (PMU) 23
4.2 Software Power Management Mechanism 27
4.2.1 Profiler 30
4.2.2 Policy Manager 32
Chapter 5 EVALUATIONS 35
5.1 Creator PXA270 Development Platform 35
5.1.1 Hardware Introduction 35
5.1.2 Evaluation Environment Setup 40
5.2 Measurement for MAR-CSE 40
5.3 Measurement for Evaluation 44
5.4 Summary 46
Chapter 6 CONCLUSION 48
References 49
Appendix A The Circuit Modification of Creator PXA270 Platform 52
Appendix B The Installation Guide of EA-DVFS Algorithm 53

[1]A. Acquaviva, L. Benini, and B. Ricco, “An adaptive algorithm for low power streaming multimedia processing.” Proc. of the Conference on Design, Automation, and Test in Europe, Mar. 2001.
[2]A. Goel, C.M. Krishna, and I. Koren. “Energy Aware Kernel for Hard Real-Time Systems.” Proc. of the 2005 International Conference on Compilers, Architectures and Synthesis for Embedded Systems, Sep. 2005.
[3]C. M. Krishna and Y.-H. Lee, “Voltage-clock-scaling Adaptive Scheduling Techniques for Low Power in Hard Real-Time Systems.” Proc. of Real-Time Technology and Applications Symposium, May 2000.
[4]C. Poellabauer and K. Schwan. “Powe-aware Video Decoding using Real-Time event Handlers.” Proc. of the 5th International Workshop on Wireless Mobile Multimedia, Sep. 2002.
[5]C. Poellabauer and K. Schwan. “Energy-Aware Media Transoconding in Wireless Systems.” Proc. of the Second IEEE International Conference on Pervasive Computing and Communications, Mar. 2004.
[6]C. Poellabauer and K. Schwan. “Energy-Aware Traffice Shaping for Wireless Real-Time Applications.” Proc. of the IEEE Real-Time and Embedded Technology and Applications Symposium, May 2004.
[7]C. Poellabauer, L. Singleton, and K. Schwan, “Feedback based dynamic voltage and frequency scaling for memory bound real-time applications.” IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), 2005, pp.234-243.
[8]Chandrakasan, A.P., Sheng, S., and Brodersen. R.W., “Low-power CMOS digital design.” IEEE Journal of Solid-State Circuits, vol. 27, April 1992, pp. 473-484.
[9]D. Grunwald et al., “Policies for Dynamic Clock Scheduling.” Proc. of the 4th Conference on Symposium on Operating System Design and Implementation, Oct. 2000.
[10]D. Qiao, S. Choi, A. Jain, and K, Shin. MiSer. “An Optimal Low-Energy Transmission Strategy for IEEE 801.11a/h.” Proc. of the ACM/IEEE International Conference on Mobile Computing and Networking, Sep. 2003.
[11]Dinesh Rajan, Russel Zuck, and Christian Poellabauer. “Workload-Aware Dual–Speed Dynamic Voltage Scaling.” Proc. of 12th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, Aug. 2006, pp.251-256.
[12]E. Bini, G. Buttazzo, and G. Lipari, “Speed Modulation in Energy-Aware Real-Time Systems.” Proc. of the 17th Euromicro Conference on Real-Time Systems, July 2005.
[13]F. Xie, M. Martonosi, and S. Malik, “Compile-time Dynamic Voltage Scaling Settings: Opportunities and Limits.” Proc. of the ACM SIGPLAN Conference on Programming Languages Design and Implementation, June 2003.
[14]F. Yao and A. Demers, “A scheduling model for reduced CPU energy.” IEEE Annual symposium on Foundations of Computer Science, 1995.
[15]H. Aydin, R. Melhem, D. Mosse, and P. M., “Power-Aware Scheduling for Periodic Real-Time Tasks.” IEEE Transactions on Computers, vol.53, no.5, May 2004.
[16]J. Pouwelse, K. Langendoen, and H. Sips, “Energy priority scheduling for variable voltage processors.” Proc. of Internantional Symposium on Low Power Electronics and Design, Aug. 2001.
[17]Jianlin Zhuo and Chaitali Chakrabarti, “An Efficient Dynamic Task Scheduling Algorithm for Battery Powered DVS Systems.” Proc. of the Asia adn South Pacific-Design Automation Conference, Jan. 2005, pp.846-849.
[18]K. Choi, R. Soma, and M. Pedram, “Dynamic Voltage Frequency Scaling based on Workload Decomposition.” Proc. of the International Symposium on Low Power Electronics and Design, Aug. 2004.
[19]K. Seth, A. Anantaraman, F. Mueller, and E. Rotenberg, “FAST: Frequency-Aware Static Timing Analysis.” Proc. of the 24th IEEE International Real-Time Systems Symposium, 2003.
[20]L. Benini, A. Bogliolo, G. Paleologo, and G.D. Micheli, “Policy optimization for dynamic power management.” IEEE Transaction on Computer-Aided Design of Integrated Circuits and Systems, vol. 18, 1999, pp.813-833.
[21]L. Benini, A. Bogliolo, and Giovanni De Micheli, “A survey of design techniques for system-level dynamic power management.” IEEE Transactions on VLSI, June 2000.
[22]M. R. Guthaus, J. S. Ringenberg, D. Ernst, T. M. Austin, T.Mudge, and R.B. Brown, “Mibench: A free commercially representative embedded benchmark suite.” IEEE International Workshop on Workload Characterizaton, 2001.
[23]M. Srivastava, A. Chandrakasan, and R. Brodersen, “Predictive system shutdown and other architectural techniques for energy efficient programmable computation.” IEEE Transaction on VLSI System, vol. 4, Mar 1996, pp.42–55.
[24]M. Weiser, B. Welch, A. Demers, and S. Shenker, “Scheduling for Reduced CPU Energy.” Proc. of the 1st Usenix Conference on Operating Systems Design & Implementation, Nov. 1994.
[25]N.C. Paver, M.H. Khan, and B.C. Aldrich, “Accelerating Mobile Multimedia Using Intel WirelessTM MMX Technology.” Proc. of IEEE 6th International Symposium on Multimedia Software Engineering, Dec 2004, pp.491-498.
[26]O.S. Unsal and I. Koren, “System-Level Power-Aware Design Techniques in Real Time Systems.” Proc. of the IEEE, Specical Issue on Real-Time Systems, Vol.91, July 2003, pp.1055-1069.
[27]P. Greenawalt, “Modeling Power Management for Hard Disks.” Proc. of International Workshop Modeling, Analysis, and Simulation for Computer and Telecomm Systems, 1994, pp.62-65.
[28]P. Mejejia-Alvarez, E. Levner, and D. Mosse, “Power-Optimize Scheduling Server for Real-Time Taks.” Proc. of the 8th IEEE Real-Time and Embedded Technology and Applications Symposium, Sep. 2002.
[29]Padmanabhan Pillai and Kang G. Shin, “Real-time dynamic voltage scaling for low-power embedded operating systems,” Proc. of 18th Symposium on Operating Systems Principles, Banff, Canada, Oct. 2001.
[30]Priya N. Vaidyna, Moinul H. Khan, Bryan Morgan, and Premanand Sakarda, “System Level Adaptive Framework for Power and Performance Scaling On Intel® PXA27X Processor.” Proc. of the 2005 IEEE International Conference on Accustics, Speech, and Signal Processing, Vol. 5, Mar. 2005, pp.657-660.
[31]Q. Wu, V.J. Reddi, Y. Wu, J. Lee, D. Connors, D. Brooks, M. Martonisi, and D. W. Clark, “A Dynamic compilation Framework for Controlling Microprocessor Energy and Performance.” Proc. of the 38th International Symp. on Microarchitecture, 2005.
[32]S. Saewong and R. R. Rajkumar, “Practical voltage-Scaling for Fixed-Priority RT-Systems.” Proc. of the the 9th IEEE Real-Time and Embedded Technology and Applications Symposium, 2003, p.106.
[33]T. Ishihara and H. Yasuura, “Voltage Scheduling Problem for Dynamically Variable Voltage Processors.” Proc. of the International Symposium on Low Power Electronics and Design, 1998.
[34]T. Pering, T. Burd, and R. Brodersen, “The simulation and evaluation of dynamic voltage scaling algorithms.” Proc. of International Symposium on Low Power Electronics and Design, Aug. 1998, pp.76-81.
[35]V. Pallipadi and A. Starikovskiy, “The Ondemand Governor.” Proc. of the Linux Symposium, vol. 2, July 2006.
[36]V. Rao, G. Singhal, and A. Kumar, “Real Time Dynamic Scaling For Embedded Systems.” Proc. of the 17th International Conference on VLSI Design, 2004.
[37]W. Kim, J. Kim, and S. Min. “A Dynamic Voltage Scaling Algorithm for Dynamic Voltage Scaling Algorithms for Hard Real-Time Systems.” Proc. of the 8th IEEE Real-Time and Embedded Technology and Applications Sysmposium, Sep. 2002.
[38]W. Kim, J. Kim, and S. Min, “A Dynamic Scaling Algorithm for Dynamic-Priority Hard Real-Time System Using Slack Time Analysis”, Proc. of the Conference on Design, Automation and Test in Europe, 2002, p.788.
[39]W. Yuan and K. Nahrstedt, “Pactical Voltage Scaling for Mobile Multimedia Devices.” Proc. of the 12th Annual ACM International Conference on Multimedia, 2004, pp.924-931.
[40]Weiser M., Welch B., Demers A., and Shenker S., “Scheduling for reduced CPU energy.” Proc. of Symposium on Operating Systems Design and Implementation, Usenix Association, Nov. 1994.
[41]Xiliang Zhong, and Cheng-Zhong Xu, “Frequency-Aware Energy Optimization for Real-Time Periodic and Aperiodic Tasks.” Proc. of the 2007 ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools, Jun. 2007.
[42]Y. Shin, K. Choi. “Power conscious fixed priority scheduling for hard real-time systems.” Proc. of the 36th ACM/IEEE Conference on Design Automation, 1999, pp.134-139.
[43]Y.Shin, K.Choi, and T. Sakurai, “Power Optimization of Real-Time Embedded Systems on Variable Speed Processors.” Proc. of the International Conference on Compuer-Aided Design, Nov. 2000, pp.365-368.
[44]Y. Zhu and F. Mueller, “Feedback EDF Scheduling Exploting Hardware-assisted Asynchronous Dynamic Voltage Scaling.” Proc. of the 2005 ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools for Embedded systems, 2005, pp.203-212.
[45] Intel XScale® Technology http://www.intel.com/design/intelxscale/
[46]“Intel PXA27x Processor Family Developer''s Manual”, http://developer.intel.com/design/pca/applicationsprocessors/manuals/280000.htm
[47]“Intel PXA270 Processor Electrical, Mechanical, and Thermal Specification”, http://www.intel.com/design/pca/applicationsprocessors/datashts/280002.htm
[48]“Intel XScale Core Developer''s Manual”, http://www.intel.com/design/intelxscale/273473.htm
[49]“Power Management in the Intel PAX27x Series Application Porcessors”, http://cache-www.intel.com/cd/00/00/21/70/217033_217033.pdf
[50] Transmeta Corporation. http://www.transmeta.com/
[51]“Wireless Intel SpeedStep Power Manager”, http://www.intel.com/design/pca/applicationsprocessors/whitepapers/300577.htm