High-level energy macromodeling of embedded software

Tat Kee Tan1, A. Raghunathan2, G. Lakshminarayana3, N.K. Jha1
1Department of Electrical Engineering, Princeton University, Princeton, NJ, USA
2NEC C and C Research Laboratories, Princeton, NJ, USA
3Alphion Corporation, Eatontown, NJ, USA

Tóm tắt

Presents an efficient and accurate high level software energy estimation methodology using the concept of characterization-based macromodeling. In characterization-based macromodeling, a function or subroutine is characterized using an accurate lower level energy model of the target processor to construct a macromodel that relates the energy consumed in the function under consideration to various parameters that can be easily observed or calculated from a high-level programming language description. The constructed macromodels eliminate the need for significantly slower instruction-level interpretation or hardware simulation that is required in conventional approaches to software energy estimation. Two different approaches to macromodeling for embedded software offer distinct efficiency-accuracy characteristics: 1) complexity-based macromodeling, where the variables that determine the algorithmic complexity of the function under consideration are used as macromodeling parameters and 2) profiling-based macromodeling, where internal profiling statistics for the functions are used as the parameters in the energy macromodels.

Từ khóa

#Embedded software #Hardware #Computer architecture #Energy consumption #Power system modeling #Computer languages #Power dissipation #Software algorithms #Statistics

Tài liệu tham khảo

sedgewick, 1998, Algorithms in C burger, 1997, The SimpleScalar tool set version 2 0 10.1109/92.335012 brooks, 2000, Wattch: a framework for architectural-level power analysis and optimizations, Proceedings of 27th International Symposium on Computer Architecture (IEEE Cat No RS00201) ISCA, 83 10.1109/LPE.1994.573194 sami, 2000, Instruction-level power estimation for embedded VLIW cores, Proceedings of the Eighth International Workshop on Hardware/Software Codesign CODES 2000 (IEEE Cat No 00TH8518) HSC, 34 10.1145/337292.337436 10.1145/335043.335044 10.1145/337292.337427 10.1109/DATE.2000.840011 10.1109/DAC.1999.782199 10.1145/337292.337786 monteiro, 1996, Computer-Aided Design Techniques for Low Power Sequential Logic Circuits 10.1109/MICRO.1997.645827 benini, 1997, Dynamic Power Management Design Techniques and CAD Tools 10.1145/143365.143490 10.1145/337292.337437 10.1109/DAC.1997.597199 10.1109/MICRO.1996.566449 10.1109/ICCD.1998.727070 chang, 2000, cycle-accurate energy consumption measurement and analysis: case study of arm7tdmi, ISLPED 00 the 2000 International Symposium on Low Power Electronics and Design (Cat No 00TH8514) LPE-00, 185, 10.1145/344166.344576 10.1007/978-1-4615-2307-9 sama, 2000, speeding up power estimation of embedded software, ISLPED 00 the 2000 International Symposium on Low Power Electronics and Design (Cat No 00TH8514) LPE-00, 191, 10.1109/LPE.2000.155276 10.1007/978-1-4615-2325-3 wolf, 2001, segment-wise timing and power measurement in software emulation, Proc DATE 2001 Designer s Forum, 165 10.1109/43.728914 10.1109/92.748204 benini, 1996, regression models for behavioral power estimation, Proc Int Workshop&#x2014 Power and Timing Modeling Optimization and Simulation 10.1145/348019.348081 myers, 1989, Classical and Modern Regression With Application 10.1109/ICVD.2001.902666