User Experience Enhanced Task Scheduling and Processor Frequency Scaling for Energy-Sensitive Mobile Devices

Abstract

Mobile devices such as smartphones, tablets and various tiny smart equipment are widely subjected to the limit of energy consumption. Moreover, user experience has become an up-to-the-moment research topic in mobile devices. In most mobile devices, the processor is usually set in a higher frequency level to provide better performance, which would cause the increment of energy consumption. In order to address this issue, in this paper, we propose a novel task scheduling model to harvest the energy saving and user experience improvement in energy-sensitive mobile devices. In our task scheduling model, we introduce the concept of user-centric task, and present tow judgmental principles to determine user-centric tasks (i.e., foreground/interactive tasks). For enhancing user experience, we reduce slack time of the user-centric tasks properly to achieve a shorter response time. We then present two algorithms to calculate the optimal frequencies and allocate the CPU resources. The experimental evaluation results show that our task scheduling model can achieve a tradeoff between energy consumption and performance, and enhance user experience.