【摘要】 功耗与计算强度成正比 ,在处理一般计算任务时 ,微处理器并不需要工作在最高的速度 ,因为高速意味高功耗。为了避免计算资源的浪费和降低功耗 ,此时应该使其在尽可能低的电压下工作。本文设计了一种基于任务驱动的低功耗方法 ,采用多个电压取代单一电压供电模式 ,根据待处理任务的实际计算需要来选定工作电压 ,既按时完成计算任务 ,又避免了功耗的浪费 ,实验表明该方法能降低 40～ 5 0 %的功耗更多还原

【Abstract】 Reduction of power dissipation in microprocessor design is becoming a key design constraint.This is motivated by the booming of portable electronics,in which battery weight and size is critical.Microprocessor has shown strong capabilities on data processing such as handwriting and speech recognition,Video encoding and decoding,etc.,but usually it targets for a lot of general purpose high latency computing tasks,in which the energy and computing resource can be wasted when it still on the top throughput.As for power dissipation is quadratic proportion to the working voltage,decrease working voltage can result in a strategy to provide orders of magnitude of power reduction,but the throughput is damaged at the same time.In this paper a task-driven energy-efficient methodology has been explored to minimize power.A Genetic Algorithm was applied to select a favorable voltage for each task base on the time limit of task queue.By introduce to this system,about 40~50% power dissipation was cut on average compared to work at a single voltage of 3.3V.更多还原