Energy consumption is one of the most important design constraints for modern microprocessors, and designers have proposed many energy-saving techniques. Looking beyond the traditional hardware low-power designs, software optimization is becoming a significant strategy for the microprocessor to lower its energy consumption. This paper describes an intra-application identification and reconfiguration mechanism for microprocessor energy reduction. Our mechanism employs a statistical sampling method during training runs to identify code sections among application that have appropriate IPC (Instructions per Cycle) values and could make contributions to program runtime energy reduction, and then profiles them to dynamically scale the voltage and frequency of the microprocessor at appropriate points during execution. In our simulation, our approach achieves energy savings by an average of 39% with minor performance degradation, compared to a processor running at a fixed voltage and speed.
