DEVELOPMENT OF SOFTWARE TOOL FOR THE ANALYSIS OF ENERGY EFFICIENCY
DOI:
https://doi.org/10.14529/cmse130203Abstract
In this paper we consider the problem of energy efficient software design. We mostly focus on programming for mobile devices. A new software toolfor the analysis of energy efficiency of applications for mobile devices is proposed. The tool allows collecting and processing experimental data, which characterizes the energy efficiency of applications which are running on the platform. We compare the similar instruments. We describe thebasic functionality, method of use and the results of use on the matrix multiplication example. The analysis of the energy efficient algorithm implementations is done by using the SSE instruction set, as well as parallel programming technologies OpenMP and Cilk Plus. Additionally, we investigate the dependence of energy efficiency on the number of used threads, the quantity of cache misses and the number of transitions between the C-States of CPU. We formulate conclusionsfor the results of the analysis and present plans to further developing.
References
Scheduling for Reduced CPU Energy / M. Weiser, B. Welch, A. Demers et al. // Proceedings of the 1st USENICS Symposium on Operating Systems Design and Implementation (Monterey, CA). – Nov. 1994. – P. 13–23.
Bekas, C. A new energy aware performance metric / C. Bekas, A. Curioni // Computer Science – R&D. – 2010. –Vol. 25. – P. 187–195.
Benini, L. Dynamic Power Management: Design Techniques and Cad Tools / L. Benini, G. De Micheli – Kluwer Academic Publishers, 1998. –231 p.
Dynamic Voltage Scaling and Power Management for Portable Systems / T. Simunic, L. Benini, A. Acquaviva et al. // Proceedings of the 38th
conference on Design automation (Las Vegas, Nevada). – Jun. 2001. – P. 524–529
Simunic, T. Energy efficient system design and utilization: PhD Thesis / T. Simunic – Stanford University (Stanford, CA), 2001. – 128 p.
Snowdon, D. Power Management and Dynamic Voltage Scaling: Myths and Facts / D. Snowdon, S. Ruocco, G. Heiser // Proceedings of the 7th ACM & IEEE International conference on Embedded software (Salzburg). – Sep. 2007. – P. 84–93.
Kappiah, N. Just In Time Dynamic Voltage Scaling: Exploiting Inter–Node Slack to Save Energy in MPI Programs / N. Kappiah, V.W. Freeh, D.K. Lowenthal // Proceedings of the ACM/IEEE SC 2005 Conference (Seattle, Washington). – Nov. 2005. – P. 33.
Reducing power with performance constraints for parallel sparse applications / G. Chen, K. Malkowski, M.T. Kandemir et al. // Proceedings of the 19th
International Parallel & Distributed Processing Symposium (Denver, Colorado). – Apr. 2005. – P. 8.
Intel Corp., Intel 64 and IA-32 Architectures Optimization Reference Manual / URL: http://www.intel.com/content/dam/doc/manual/64-ia-32-architectures-optimization-manual.pdf (дата обращения: 10.1.2013), Chapter 11. P. 423–430.
Energy-Efficient Platforms – Considerations for Application Software and Services / URL: http://download.intel.com/technology/pdf/322304.pdf (дата обращения: 05.09.2012).
Tiwari, V. Power Analysis of Embedded Software: A First Step / V. Tiwari, S. Malik, A. Wolfe // IEEE Transactions on VLSI Systems. – 1994. –Vol. 2, No. 4. – P. 437–445.
Instruction Level Power Analysis and Optimization of Software / V. Tiwari, S. Malik,
A. Wolfe et al. // Proceedings of the 9th International Conference VLSI Design (Bangalore, India). – Jan. 1996. – P. 326–328.
Instruction scheduling for power reduction in processor–based system design / H.H. Tomiyama, T. Ishihara, A. Inoue et al. // Proceedingsof the Conference Design, Automation ant Test in Europe (Paris, France). – Feb. 1998. – P. 855–860.
Simunic, T. Energy–Efficient Design of Battery–Powered Embedded Systems / T. Simunic, L. Benini, G. De Micheli // IEEE Transactions on VLSI Systems. – 2001. –Vol. 9, No. 1. – P. 15–28.
The impact of source code transformations on software power and energy consumption / C. Brandoles, W. Fornaciari, F. Salice et al. // World Scientific Journal of Circuits Systems and Computers. – 2002. – Vol. 11, No. 5 – P. 477–502.
Power Checker Web Site / URL: http://software.intel.com/en-us/blogs/2011/06/27/intel-power-checker/ (дата обращения: 10.10.2012).
Power Informer Web Site / URL: http://software.intel.com/en-us/articles/intel-powerinformer/ (дата обращения: 10.10.2012).
Joulemeter Web Site / URL: http://research.microsoft.com/en-us/downloads/fe9e10c5-5c5b-450c-a674-daf55565f794/ (дата обращения: 10.10.2012).
Perfmon Web Site / URL: http://technet.microsoft.com/en-us/library/bb490957.aspx (дата обращения: 10.10.2010).
Power Checker Web Site / URL: https://01.org/powertop/ (дата обращения: 10.10.2012).
