Optimisation of Energy Consumption of Soft Real-Time Applications by Workload Prediction

Embedded real-time systems often operate under energy constraints due to a limited battery lifetime. Modern processors provide techniques for dynamic voltage and frequency scaling to reduce energy consumption. However, while the processor possibly operates at a lower clock frequency, the running applications should still meet their deadlines and thus set some limits to the use of scaling techniques. In this paper, we propose auto correlation clustering (ACC) as a technique to predict the workload of single iterations of a periodic soft real-time application. Based on this prediction we adjust the processor performance such that deadlines are exactly met. We compare our technique to the broadly implemented race-to-idle (RTI) and identify situations where ACC can gain higher energy savings than RTI. Additionally, ACC can help saving energy in multithreaded processors where RTI can be applied only with a high overhead if at all.