Minimizing the power dissipation of electronic systems is one of the most critical areas of research today. Many techniques exist to understand and control power leakage in the hardware itself yet there has been less effort towards understanding the effect of software on power consumption for the overall system. While processor data sheets and programmer's guides may contain hundreds of pages, they typically list only halted, typical and maximum power dissipation figures. Even if per-instruction power consumption data were made available for the CPU itself, it cannot account for interactions with off-chip memory and peripherals, which is specific to the target and the software running on it. This paper describes a simple technique that establishes a baseline for measuring the effect of running software on an arbitrary target with unmodified hardware, using only basic lab equipment
[1]
Mike Tien-Chien Lee,et al.
Power analysis and low-power scheduling techniques for embedded DSP software
,
1995
.
[2]
Chandra Krintz,et al.
Predicting Program Power Consumption
,
2002
.
[3]
Sharad Malik,et al.
Power analysis and minimization techniques for embedded DSP software
,
1997,
IEEE Trans. Very Large Scale Integr. Syst..
[4]
Kwanho Kim,et al.
Real-time per-cycle energy consumption measurement of digital systems
,
2000
.
[5]
Sang Lyul Min,et al.
An Accurate Instruction-Level Energy Consumption Model for Embedded RISC Processors
,
2001
.
[6]
Sharad Malik,et al.
Power analysis of embedded software: a first step towards software power minimization
,
1994,
IEEE Trans. Very Large Scale Integr. Syst..