臺灣博碩士論文加值系統

和同質多核心處理器比較起來，異質多核心處理器已經被
提出作為一個比較好的折衷方案在於效能與能源消耗方面。在
此同時也浮現了新的挑戰；那就是執行緒與核心之間的配對問
題。 本論文提出一個基於模糊控制的排程器；在我們的研究
中，我們架構了一個相同指令集架構但是不同核心有不同的頻
率的異質多核心平台。我們所提出的模糊控制排程器週期性地
收集執行緒的程式特性，並且根據這些特性來決定該執行緒應
該被分配到哪一個核心上。實驗結果顯示，我們的模糊控制排
程器對於記憶體存取密集的程式可以節省相當多能源以及得
到顯著的能源使用效率而效能並未犧牲很多。

Asymmetric multicore processors have been proposed
as a better trade-off between performance and power
consumption over symmetric multicore processors. They
also reveal challenges in mapping threads to cores. We
propose a new scheduler based on fuzzy control theory.
In this work, we configure an asymmetric multicore
system in which cores share the same ISA but run at
different frequencies. Our fuzzy scheduler decides
thread-to-core assignment based on periodical run-time
performance characteristics. Evaluation results
demonstrate that our fuzzy scheduler saves significant
energy and achieves better energy-delay product for
memory-intensive programs while sacrificing
performance slightly.

X ii
Abstract iii
Πiv
List of Figures vii
List of Tables ix
1 Introduction 1
1.1 Asymmetric Multicore Systems . . . . . . . . . . . . . . . . . 1
1.2 Toward Better Utilization of AMPs . . . . . . . . . . . . . . . 3
2 Related work 5
3 Implementation 8
3.1 JVM and JVM Tool Interface . . . . . . . . . . . . . . . . . . 8
3.2 Interesting Performance Metrics . . . . . . . . . . . . . . . . . 9
3.3 Thread Assignment Policy . . . .. . . . . . . . . . . . . . . . 10
3.4 Fuzzy Scheduler . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4 Experiment Setup 17
4.1 System Conguration . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Benchmark Conguration . . . . . . . . . . . . . . . . . . . . 19
4.3 Power and Energy Measurement . . . . . . . . . . . . . . . . . 20
5 Evaluation 21
5.1 Benchmark Characteristics . . . . . . . . . . . . . . . . . . . . 21
5.2 Experiment Results . . . . . . . . . . . . . . . . . . . . . . . . 23
6 Conclusion and Future works 30
6.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.2 Future works . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Bibliography 32

[1] EPS12V Power Supply Design Guide - A Server System Infrastructure
(SSI) Specication For Entry Chassis Power Supplies, 2.91 edition.
[2] Jvm tool interface.
[3] Performance counters on linux.
[4] Project sikuli.
[5] Specjvm2008.
[6] Michela Becchi and Patrick Crowley. Dynamic thread assignment on
heterogeneous multiprocessor architectures. In Proceedings of the 3rd
conference on Computing frontiers, CF '06, pages 29{40, New York,
NY, USA, 2006. ACM.
[7] W. L. Bircher, Jason Law, Madhavi Valluri, and Lizy K. John. Effective use of performance monitoring counters for run-time prediction of power. Technical report, nov. 2004.
[8] J. Mark Bull, L. A. Smith, M. D. Westhead, D. S. Henty, and R. A.
Davey. A benchmark suite for high performance java. Concurrency -
Practice and Experience, pages 375{388, 2000.
[9] Didier Dubois. Fuzzy sets and their applications : Vilem novak, translated from czechoslovakian. bristol and philadelphia: Adam hilger, 1989, 248 pages. Mathematical Social Sciences, 21(2):193{197, April 1991.
[10] Alexandra Fedorova, Juan Carlos Saez, Daniel Shelepov, and Manuel
Prieto. Maximizing power eciency with asymmetric multicore systems.
Commun. ACM, 52(12):48{57, December 2009.
[11] Matt Gillespie. Preparing for the second stage of multi-core hardware: Asymmetric (heterogeneous) cores. Technical report, Intel, 2008.
[12] Peter Greenhalgh. Big.little processing with arm cortex-a15 & cortex-a7. White paper, ARM, 2011.
[13] John Hennessy, John L. Hennessy, David Goldberg, and David A. Patterson. Computer Architecture: A Quantitative Approach. Morgan Kaufmann Publishers, 4rd edition.
[14] Intel Corporation. Intel 64 and IA-32 Architectures Software Developer's Manual - Volume 3B: System Programming Guide, May 2011. chapter 30 Performance monitoring.
[15] R. Kumar, K.I. Farkas, N.P. Jouppi, P. Ranganathan, and D.M. Tullsen. Single-isa heterogeneous multi-core architectures: the potential for processor power reduction. In Microarchitecture, 2003. MICRO-36. Proceedings. 36th Annual IEEE/ACM International Symposium on, pages 81 ~ 92, dec. 2003.
[16] R. Kumar, D.M. Tullsen, N.P. Jouppi, and P. Ranganathan. Heterogeneous chip multiprocessors. Computer, 38(11):32 { 38, nov. 2005.
[17] Rakesh Kumar, Dean M. Tullsen, Parthasarathy Ranganathan, Nor-
man P. Jouppi, and Keith I. Farkas. Single-isa heterogeneous multi-core architectures for multithreaded workload performance. SIGARCH Comput. Archit. News, 32(2):64{, March 2004.
[18] Tong Li, Dan Baumberger, David A. Koufaty, and Scott Hahn. Ef-
cient operating system scheduling for performance-asymmetric multi-
core architectures. In Proceedings of the 2007 ACM/IEEE conference on
Supercomputing, SC '07, pages 53:1{53:11, New York, NY, USA, 2007.
ACM.
[19] Tong Li, P. Brett, R. Knauerhase, D. Koufaty, D. Reddy, and S. Hahn.
Operating system support for overlapping-isa heterogeneous multi-core
architectures. In High Performance Computer Architecture (HPCA),
2010 IEEE 16th International Symposium on, pages 1 {12, jan. 2010.
[20] Tong Li, Paul Brett, Barbara Hohlt, Rob Knauerhase, Sean D.
McElderry, and Scott Hahn. Operating system support for shared-
isa asymmetric multi-core architectures. In Proceedings of the Fourth
Annual Workshop on the Interaction between Operating Systems and
Computer Architecture (WIOSCA '08), pages 19{26, June 2008.
[21] J.C. Mogul, J. Mudigonda, N. Binkert, P. Ranganathan, and V. Talwar.
Using asymmetric single-isa cmps to save energy on operating systems.
Micro, IEEE, 28(3):26 {41, may-june 2008.
[22] Priya Nagpurkar, Chandra Krintz, Michael Hind, Peter F. Sweeney,
and V. T. Rajan. Online phase detection algorithms. In Proceedings
of the International Symposium on Code Generation and Optimization,
CGO '06, pages 111{123, Washington, DC, USA, 2006. IEEE Computer
Society.
[23] H.R. Pourshaghaghi and J.P. de Gyvez. Dynamic voltage scaling based on supply current tracking using fuzzy logic controller. In Electronics, Circuits, and Systems, 2009. ICECS 2009. 16th IEEE International Conference on, pages 779 {782, dec. 2009.
[24] Dheeraj Reddy, David Koufaty, Paul Brett, and Scott Hahn. Bridging functional heterogeneity in multicore architectures. SIGOPS Oper. Syst. Rev., 45(1):21{33, February 2011.
[25] Daniel Shelepov, Juan Carlos Saez Alcaide, Stacey Jeery, Alexan-
dra Fedorova, Nestor Perez, Zhi Feng Huang, Sergey Blagodurov, and
Viren Kumar. Hass: A scheduler for heterogeneous multicore systems.
SIGOPS Oper. Syst. Rev., 43:66{75, April 2009.
[26] T. Sondag and H. Rajan. Phase-based tuning for better utilization
of performance-asymmetric multicore processors. In Code Generation
and Optimization (CGO), 2011 9th Annual IEEE/ACM International
Symposium on, pages 11 {20, april 2011.
[27] Hsin-Ching Sun, Bor-Yeh Shen, Wuu Yang, and Jenq-Kuen Lee. Migrating java threads with fuzzy control on asymmetric multicore systems for better energy delay product. In International Conference on Computing and Security, Ulaanbaatar, Mongolia, July 2011.
[28] Qiming Teng, P.F. Sweeney, and E. Duesterwald. Understanding the
cost of thread migration for multi-threaded java applications running on a multicore platform. In Performance Analysis of Systems and Software, 2009. ISPASS 2009. IEEE International Symposium on, pages 123 ~ 132, april 2009.
[29] Viswanath Krishnamurthy Tyler Sondag and Hridesh Rajan. Predictive thread-to-core assignment on a heterogeneous multi-core processor. In PLOS '07: ACM SIGOPS 4th Workshop on Programming Languages and Operating Systems, October 2007.
[30] Frederik Vandeputte, Lieven Eeckhout, and Koen De Bosschere.
Exploiting program phase behavior for energy reduction on multi-
conguration processors. J. Syst. Archit., 53(8):489{500, August 2007.
[31] Lot A. Zadeh. Fuzzy logic, neural networks, and soft computing. Commun. ACM, 37(3):77{84, March 1994.