Adaptive Approach in Handling Human Inactivity in Computer Power Management

Ria Candrawati, Nor Laily Hashim

Abstract


Human inactivity is handled by adapting the behavioral changes of the users. Human inactivity refers to as unpredictable workload of a complex system that is caused by increments of amount in power consumption and it can be handled automatically without the need to set a fixed time for changing the computer state. This is happens due to lack of knowledge in a software system and the software self-adaptation is one approach in dealing with this source of uncertainty. This paper observes human inactivity and Power management policy through the application of reinforcement learning approach in the computer usage and finds that computer power usage can be reduced if the idle period can be intelligently sensed from the user activities. This study introduces Control, Learn and Knowledge model that adapts the Monitor, Analyze, Planning, Execute control loop integrates with Q Learning algorithm to learn human inactivity period to minimize the computer power consumption. An experiment to evaluate this model was conducted using three case studies with same activities. The result show that the proposed model obtained those 5 out of 12 activities shows the power decreasing compared to others

Keywords


Reinforcement Learning; Dynamic Power Management; Human Inactivity;

Full Text:

PDF

References


Gupta, P. K. and Singh, G., "Minimizing Power Consumption by Personal Computers: A Technical Survey," Int. J. Inf. Technol. Comput. Sci. 4(10):57–66, 2012.

Irani, S. Shukla, S. and Gupta, R., "Online strategies for dynamic power management in systems with multiple power-saving states," ACM Trans. Embed. Comput. Syst. 2(3):325–346, 2003.

Sandhu, S. Rawal, A. Kaur, P. and Gupta, N., "Major components associated with green networking in information communication technology systems,” Commun. Appl. (ICCCA), 2012 Int. Conf. Comput.1–6, 2012.

Esfahani, N. and Malek, S. Uncertainty in self-adaptive software systems, Softw. Eng. Self-Adaptive Syst. II, 2013.

Shen, H. Tan, Y. Lu, J. Wu, Q.and Qiu, Q., “Achieving autonomous power management using reinforcement learning,” ACM Trans. Des. Autom. Electron. Syst.18(2):1–32, 2013.

Candrawati, R. Hashim, N. L. Mahmuddin, M. and Irwan, H., "A Model of Framework of Control , Learn , and Knowledge for Computer Power Management" In Proceeding of Knowledge Management International Conference, 2014.

Schumann, M. a. Drusinsky, D. Michael, J. B.and Wijesekera, D.,"Modeling human-in-the-loop security analysis and decision-making processes," IEEE Trans. Softw. Eng. 40(2):154–166, 2014.

Kothari, S. Deepak, A. Tamrawi, A. Holland, B. and Krishnan, S., "A ‘ Human-in-the-loop ’ Approach for Resolving Complex Software

Anomalies," in IEEE International Conference on Systems, Man, and Cybernetics, 2014,1971–1978, 2014.

Munir, S. Stankovic, J. Liang C., and Lin, S. Reducing Energy Waste for Computers by Human-in-the-Loop Control, 2013.

IBM. An architectural blueprint for autonomic computing, 2005.

Lemos, R. De Giese, H. and Müller, H., "Software engineering for selfadaptive systems: A second research roadmap," Softw. Eng.1–32, 2013.

Brun, Y. Di, G. Serugendo, M. Gacek, C. Giese, H.and Shaw, M. Engineering Self-Adaptive Systems through Feedback Loops.. 48–70, 2009.

Chedid W. and Yu, C. Survey on power management techniques for energy efficient computer systems, 2002.

Lu Y.-H. and Micheli, G. De, "Comparing system level power

management policies," IEEE Des. Test Comput. 18(2):10–19, 2001.[15]

Srivastava, M. B. Chandrakasan, A. P. and Brodersen, R. W.,

"Predictive System Shutdown and Other Architectural Techniques for Energy Efficient Programmable Computation," IEEE Trans. VERY

LARGE SCALE Integr. Syst. 4(I):42–55, 1996.

Benini, L. Bogliolo, A. and De Micheli, G., "A survey of design techniques for system-level dynamic power management," IEEE Trans. Very Large Scale Integr. Syst., 8(3):299–316, 2000.

Barto, A. and Dietterich, T. Reinforcement learning and its relationship to supervised learning, Handb. Learn. Approx. Dyn. Program, 2004.

Gurumurthi, S. Sivasubramaniam, A. Irwin, M. J. Vijaykrishnan, N. and Kandemir, M. Using complete machine simulation for software power estimation: The softwatt approach, High-Performance Comput. Arch, 2002.

Tan, Y. and Qiu, Q., "A Framework of Stochastic Power

Management Using Hidden Markov Model," 92–97, 2008.

Dhiman, G. and Rosing, T., "Dynamic Power Management Using Machine Learning," in 2006 IEEE/ACM International Conference on Computer Aided Design, 2006. 747–754, 2006.

Khan U. A.and Rinner, B., "A Reinforcement Learning Framework for Dynamic Power Management of a Portable, Multi-camera Traffic

Monitoring System," IEEE Int. Conf. Green Comput. Commun. Nov.

557–564, 2012.

Khan, U.and Rinner, B., "Online learning of timeout policies for dynamic power management," ACM Trans. Embed. Comput. 13(4), 2014.

Higgs, T. Energy Efficient Computing, IEEE,210–215, 2007.

Moshnyaga, V. G. How to Really Save Computer Energy?, Computer (Long. Beach. Calif).3, 2008.

Moshnyaga, V. G. he use of eye tracking for PC energy management, Proc. 2010 Symp. Eye-Tracking Res. Appl. - ETRA ’10, 113, 2010.

CPUID, “HWMONITOR-PRO,” 2016. [Online]. Available:

http://www.cpuid.com/softwares/hwmonitor-pro.html. [Accessed: 31-Dec-2015].


Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

ISSN: 2180-1843

eISSN: 2289-8131