research-article

Supervised learning based power management for multicore processors

Authors:

Massoud PedramAuthors Info & Claims

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Volume 29, Issue 9

Pages 1395 - 1408

https://doi.org/10.1109/TCAD.2010.2059270

Published: 01 September 2010 Publication History

Abstract

This paper presents a supervised learning based power management framework for a multi-processor system, where a power manager (PM) learns to predict the system performance state from some readily available input features (such as the occupancy state of a global service queue) and then uses this predicted state to look up the optimal power management action (e.g., voltage-frequency setting) from a precomputed policy table. The motivation for utilizing supervised learning in the form of a Bayesian classifier is to reduce the overhead of the PM which has to repetitively determine and assign voltage-frequency settings for each processor core in the system. Experimental results demonstrate that the proposed supervised learning based power management technique ensures system-wide energy savings under rapidly and widely varying workloads.

References

[1]

D. Li, Q. Xie, and P. H. Chou, "Scalable modeling and optimization of mode transitions based on decoupled power management architecture," in Proc. Design Autom. Conf., Jun. 2003, pp. 119-124.

Digital Library

[2]

Y.-H. Lu and G. De. Micheli, "Comparing system-level power management policies," IEEE Design Test Comput., vol. 18, no. 2, pp. 10-19, Mar.-Apr. 2001.

Digital Library

[3]

A. Naveh, E. Rotem, A. Mendelson, S. Gochman, R. Chabukswar, K. Krishnan, and A. Kumar, "Power and thermal management in Intel core duo processor," Intel Technol. J., vol. 10, no. 2, pp. 109-122, May 2006.

[4]

H. Jung and M. Pedram, "Continuous frequency adjustment technique based on dynamic workload prediction," in Proc. Int. Conf. VLSI Design, Jan. 2008, pp. 415-420.

Digital Library

[5]

A. Iyer and D. Marculescu, "Power efficiency of voltage scaling in multiple clock, multiple voltage cores," in Proc. Int. Conf. Comput.- Aided Design, Nov. 2002, pp. 379-386.

Digital Library

[6]

R. Kumar, K. Farkas, N. Jouppi, P. Ranganathan, and D. Tullsen, "Single ISA heterogeneous multicore architecture: The potential for processor power reduction," in Proc. Symp. Microarchitecture, Dec. 2003, pp. 81- 93.

Digital Library

[7]

Q. Wu, P. Juang, M. Martonosi, and D. W. Clark, "Voltage and frequency control with adaptive reaction time in multiple-clock domain processors," in Proc. Symp. High-Performance Comput. Architecture, Feb. 2005, pp. 178-189.

Digital Library

[8]

Y. Hotta, M. Sate, H. Kimura, S. Matsuoka, T. Boku, and D. Takahashi, "Profile-based optimization of power performance by using dynamic voltage scaling of a PC cluster," in Proc. Parallel Distributed Process. Symp., Apr. 2006, pp. 8-16.

Digital Library

[9]

B. Mochocki, D. Rajan, X. S. Hu, C. Poellabacer, K. Otten, and T. Chantern, "Network-aware dynamic voltage and frequency scaling," in Proc. Real Time Embedded Technol. Applicat. Symp., Apr. 2007, pp. 215-224.

Digital Library

[10]

E. Chung, L. Benini, and G. De. Micheli, "Dynamic power management using adaptive learning tree," in Proc. Int. Conf. Comput.-Aided Design, Nov. 1999, pp. 274-279.

Digital Library

[11]

O. Chapelle, B. Scholkopf, and A. Zien, Semi-Supervised Learning. Cambridge, MA: MIT Press, 2006.

Digital Library

[12]

S. Ma and C. Ji, "Performance and efficiency: Recent advances in supervised learning," Proc. IEEE, vol. 87, no. 9, pp. 1519-1535, Sep. 1999.

[13]

P. Teich, "Multicore processor technology: Maximizing CPU performance in a power-constrained world," in Proc. Presentation Slide Microsoft Windows Hardw. Eng. Conf., Apr. 2005, pp. 7-9.

[14]

H. Zhong, S. A. Lieberman, and S. A. Mahke, "Extending multicore architectures to exploit hybrid parallelism in single-thread applications," in Proc. Int. Symp. High Performance Comput. Architecture, Mar. 2007, pp. 25-36.

Digital Library

[15]

V. Paxson, R. Sommer, and N. Weaver, "An architecture for exploiting multi-core processors to parallelize network intrusion prevention," in Proc. IEEE Sarnoff Symp., Apr. 2007, pp. 1-7.

[16]

T. D. Burd and R. W. Brodersen, "Design issues for dynamic voltage scalmg, in Proc. Int. Symp. Low Power Electron. Design, Aug. 2000, pp. 9-14.

Digital Library

[17]

A. Mittal and A. Kassim, Bayesian Network Technologies: Applications and Graphical Models. Hershey, PA: IGI Publishing, 2007.

Digital Library

[18]

T. Mitchell, Machine Learning. New York: McGraw-Hill, 1997.

Digital Library

[19]

K. Huang, Z. Xu, I. King, M. R. Lyu, and Z. Zhou, "A novel discriminative naive Bayesian network for classification," in Bayesian Network Technologies: Applications and Graphical Models, A. Mittal, A. Kassim, and T. Tan, Eds. Hershey, PA: IGI Publishing, Mar. 2007, pp. 1-12.

[20]

V. N. Vapnik, The Nature of Statistical Learning Theory, 2nd ed. New York: Springer-Verlag, 1999.

Digital Library

[21]

G. Theocharous, S. Mannor, N. Shah, P. Gandhi, B. Kveton, S. Siddiqi, and C. Yu, "Machine learning for adaptive power management," Intel Technol. J., vol. 10, no. 4, pp. 299-310, Jul. 2006.

[22]

G. Dhiman and T. S. Rosing, "Dynamic power management using machine learning," in Proc. Int. Conf. Comput.-Aided Design, Nov. 2006, pp. 747-754.

Digital Library

[23]

G. Dhiman and T. S. Rosing, "Dynamic voltage frequency scaling for multi-tasking systems using online learning," in Proc. Int. Symp. Low Power Electron. Design, Jul. 2007, pp. 207-212.

Digital Library

[24]

A. Weissel and F. Bellosa, "Self-learning hard disk power management for mobile devices," in Proc. Int. Workshop Softw. Support Portable Storage, Oct. 2006, pp. 33-40.

[25]

C. Rusu, N. AbouGhazaleh, A. Ferreira, R. Xu, B. Childers, R. Melhem, and D. Mosse, "Integrated CPU and L2 cache frequency/voltage scaling using supervised learning;' in Proc. Workshop Statistic. Mach. Learning Approaches Appl. Architectures Compilation, Jul. 2007, pp. 41-50.

[26]

W. Cohen, "Fast effective rule induction," in Proc. 12th Int. Conf. Mach. Learning, Dec. 1995. pp. 115-123.

[27]

R. Quinlan, C4.5: Program for Machine Learning. San Francisco, CA: Morgan Kaufmann, 1993.

Digital Library

[28]

A. Statnikov, C. F. Aliferis, I. Tsamardinos, D. P. Hardin, and S. Levy, "A comprehensive evaluation of multicategory classification methods for micro array gene expression cancer diagnosis," Bioinformatics, vol. 21, no. 5, pp. 631-643, 2004.

Digital Library

[29]

C. Cortes and V. Vapnik, "Support-vector networks," J. Mach. Learning, vol. 20, no. 3, pp. 273-297, 1995.

Digital Library

[30]

R. Knauerhase, P. Brett, T. Li, B. Hohlt, and S. Hahn, "Using OS observations to improve performance in multi-core systems," Proc. IEEE Micro, vol. 28, no. 3, pp. 54-66, May-Jun. 2008.

Digital Library

[31]

L. Zhao, R. Iyer, R. Illikkal, J. Moses, S. Makineni, and D. Newell, "CacheScouts: Fine-grain monitoring of shared caches in CMP platforms," in Proc. Int. Conf. Parallel Architecture Compilation Tech., Oct. 2007, pp. 339-349.

Digital Library

[32]

M. L. Seltzer, B. Raj, and R. M. Stern, "A Bayesian classifier for spectrographic mask estimation for missing feature speech recognition," J. Speech Commun., vol. 43, pp. 379-393, Mar. 2004.

[33]

A. Dempster, N. Laird, and D. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. Royal Stat. Soc. Series B, vol. 39, no. 1, pp. 1-38, 1977.

[34]

J. A. Bilmes, "A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov model," Univ. California, Berkeley, Tech. Rep. TR-97-021, 1998.

[35]

R. A. Fisher, Statistical Methods and Scientific Inference. New York: Macmillan, 1973.

[36]

A. Gosavi, Simulation-Based Optimization: Parameter Optimization Techniques and Reiriforcement Learning. Boston, MA: Kluwer Academic, 2003.

Digital Library

[37]

R. E. Bellman, Dynamic Programming. Englewood Cliffs, NJ: Princeton University Press, 1957.

Digital Library

[38]

R. J. Williams and L. C. Baird, "Tight performance bounds on greedy policies based on Imperfect value functions," Northeastern Univ., Boston, MA, Tech. Rep. NU-CCS-93-14, Nov. 1993.

[39]

K. Salchow. "Load balancing 101." White Papers from F5, Inc. {Online}. Available: http://www.f5.com/pdf/white-papers/evolution-adc-wp.pdf, http://www.theacademy.ca/load-balancing101-wp.pdf

[40]

White Paper. (2004, Apr.). "Scalable networking: Eliminating the receive processmg bottleneck--mtroduction RSS." WinHEC 2004 Version {Online}. Available: http://www.microsoft.com/whdc

[41]

IEEE 802.3 Ethernet Document {Online}. Available: http://www.ieee802. org

[42]

H. Jung and M. Pedram: A unified framework for system-level design: Modeling and performance optimization of scalable networking systems, in Proc. Int. Symp. Quality Electron. Designs, Mar. 2007, pp. 198-203.

Digital Library

[43]

Opencore. (2009). OpenRISC Processor {Online}. Available: http:// www.opencores.org

[44]

Synopsys. (2009). Synopsys Compiler {Online}. Available: http:// www.synopsys.com

[45]

B. Calhoun, J. Kao, and A. Chandrakasan, Leakage in Nanometer CMOS Technologies. Berlin, Germany: Springer, 2006.

[46]

White Paper. (2008, Sep.). Bi-Directional Current/Power Monitor with I²C Interface {Online}. Available: http://focus.ti.com

[47]

V. Srinivasan, D. Brooks, M. Gschwind, and P. Bose, "Optimizing pipelines for power and performance," in Proc. Int. Symp. Microarchitecture, Nov. 2002, pp. 333-344.

Digital Library

[48]

V. Pronk, S. V. R. Gutta, and W. F. J. Verhaegh, "Incorporating confidence in a naive Bayesian classifier," in Lecture Notes in Computer Science: User Modeling. Berlin, Germany: Springer-Verlag, Aug. 2005, pp. 317-326.

Digital Library

[49]

G. M. Foody, A. Mathur, C. Sanchez-Hernandez, and D. S. Boyd, "Training set size requirements for the classification of a specific class," J. Remote Sensing Environment, vol. 104, no. 1, pp. 1-14, Sep. 2006.

[50]

C. McNairy and R. Bhaita, "Montecito: A dual-core, dual-thread titanium processor," IEEE Micro, vol. 25, no. 2, pp. 10-20, Mar.-Apr. 2005.

Digital Library

[51]

S. Chaudhuri and V. Narasayya, "An efficient cost-driven index tuning wizard for microsoft SQL server," in Proc. 23rd Int. Conf. Very Large Databases, Sep. 1997, pp. 146-155.

Digital Library

[52]

I. Ahmand, "Easy and efficient disk I/O workload characterization in VMware ESX server," in Proc. Int. Symp. Workload Characterization, Sep. 2007, pp. 149-158.

Digital Library

[53]

E. Castro-Leon, S. Nayak, and D. Shenkar. (2009, Jul.). Data Center Powa and Thermal Management Using Intel Data Center Manager Sofollare Development Kit {Online}. Available: http://software.intel.com

[54]

Q. Qiu and M. Pedram, "Dynamic power management based on continuous time Markov decision process," in Proc. Design Autom. Conf., Jun. 1999, pp. 555-561.

Digital Library

[55]

C. Isci, A. Buyuktosunoglu, C. Cher, P. Bose, and M. Martonosi, "An analysis of efficient multicore power management policies: Maximizing performance for a given power budget," in Proc. Int. Symp. Microarchitecture, 2006, pp. 347-358.

Digital Library

Cited By

Dong PDang KNguyen DTran X(2024) A light-weight neuromorphic controlling clock gating based multi-core cryptography platform Microprocessors & Microsystems10.1016/j.micpro.2024.105040106:COnline publication date: 1-Apr-2024
https://dl.acm.org/doi/10.1016/j.micpro.2024.105040
Zhou TLin M(2023)CPU frequency scheduling of real-time applications on embedded devices with temporal encoding-based deep reinforcement learningJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2023.102955142:COnline publication date: 1-Sep-2023
https://dl.acm.org/doi/10.1016/j.sysarc.2023.102955
Hoffmann JFröhlich A(2022)Online Machine Learning for Energy-Aware Multicore Real-Time Embedded SystemsIEEE Transactions on Computers10.1109/TC.2021.305607071:2(493-505)Online publication date: 1-Feb-2022
https://dl.acm.org/doi/10.1109/TC.2021.3056070
Show More Cited By

Recommendations

Reinforcement learning based dynamic power management with a hybrid power supply
ICCD '12: Proceedings of the 2012 IEEE 30th International Conference on Computer Design (ICCD 2012)

Dynamic power management (DPM) in battery-powered mobile systems attempts to achieve higher energy efficiency by selectively setting idle components to a sleep state. However, re-activating these components at a later time consumes a large amount of ...
Hybrid power management in real time embedded systems: an interplay of DVFS and DPM techniques

Energy-aware scheduling of real time applications over multiprocessor systems is considered in this paper. Early research reports that while various energy-saving policies, for instance Dynamic Power Management (DPM) and Dynamic Voltage & Frequency ...
Dynamic Power Management in Large Manycore Systems: A Learning-to-Search Framework
The complexity of manycore System-on-chips (SoCs) is growing faster than our ability to manage them to reduce the overall energy consumption. Further, as SoC design moves toward three-dimensional (3D) architectures, the core's power density increases ...

Comments

Information & Contributors

Information

Published In

cover image IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems Volume 29, Issue 9

September 2010

153 pages

ISSN:0278-0070

Issue’s Table of Contents

Copyright © 2010.

Publisher

IEEE Press

Publication History

Published: 01 September 2010

Revised: 03 June 2009

Received: 11 November 2008

Author Tags

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

29
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 20 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Dong PDang KNguyen DTran X(2024) A light-weight neuromorphic controlling clock gating based multi-core cryptography platform Microprocessors & Microsystems10.1016/j.micpro.2024.105040106:COnline publication date: 1-Apr-2024
https://dl.acm.org/doi/10.1016/j.micpro.2024.105040
Zhou TLin M(2023)CPU frequency scheduling of real-time applications on embedded devices with temporal encoding-based deep reinforcement learningJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2023.102955142:COnline publication date: 1-Sep-2023
https://dl.acm.org/doi/10.1016/j.sysarc.2023.102955
Hoffmann JFröhlich A(2022)Online Machine Learning for Energy-Aware Multicore Real-Time Embedded SystemsIEEE Transactions on Computers10.1109/TC.2021.305607071:2(493-505)Online publication date: 1-Feb-2022
https://dl.acm.org/doi/10.1109/TC.2021.3056070
Gupta MBhargava LIndu S(2022)Deep neural network learning for power limited heterogeneous system with workload classificationComputing10.1007/s00607-021-01018-5104:1(95-122)Online publication date: 1-Jan-2022
https://dl.acm.org/doi/10.1007/s00607-021-01018-5
Zou AGarimella KLee BGill CZhang XSchlichtmann UGal RAmrouch HLi H(2020)F-LEMMAProceedings of the 2020 ACM/IEEE Workshop on Machine Learning for CAD10.1145/3380446.3430630(43-48)Online publication date: 16-Nov-2020
https://dl.acm.org/doi/10.1145/3380446.3430630
Pagani SManoj PJantsch AHenkel J(2020)Machine Learning for Power, Energy, and Thermal Management on Multicore Processors: A SurveyIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems10.1109/TCAD.2018.287816839:1(101-116)Online publication date: 1-Jan-2020
https://dl.acm.org/doi/10.1109/TCAD.2018.2878168
Gajaria DAdegbija T(2019)ARCProceedings of the International Symposium on Memory Systems10.1145/3357526.3357553(439-450)Online publication date: 30-Sep-2019
https://dl.acm.org/doi/10.1145/3357526.3357553
Dinakarrao SJoseph AHaridass AShafique MHenkel JHomayoun H(2019)Application and Thermal-reliability-aware Reinforcement Learning Based Multi-core Power ManagementACM Journal on Emerging Technologies in Computing Systems10.1145/332305515:4(1-19)Online publication date: 10-Oct-2019
https://dl.acm.org/doi/10.1145/3323055
Koo KGovindarasu MTian J(2019)Event prediction algorithm using neural networks for the power management system of electric vehiclesApplied Soft Computing10.1016/j.asoc.2019.10570984:COnline publication date: 1-Nov-2019
https://dl.acm.org/doi/10.1016/j.asoc.2019.105709
Kim RDoppa JPande P(2018)Machine Learning for Design Space Exploration and Optimization of Manycore Systems2018 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)10.1145/3240765.3243483(1-6)Online publication date: 5-Nov-2018
https://dl.acm.org/doi/10.1145/3240765.3243483
Show More Cited By

View Options

View options

Media

Figures

Other

Tables

View Issue’s Table of Contents