Energy-Aware Optimal Task Assignment for Mobile Heterogeneous Embedded Systems in Cloud Computing

Abstract

Recent quick expansions of mobile heterogeneous embedded systems have led to a remarkable hardware upgrade that support multiple core processors. The energy consumption is becoming greater along with the computation capacity grows. Cloud computing is considered one of the solutions to mitigating energy costs. However, the simply offloading the computations to the remote side cannot efficiently reduce the energy consumptions when the energy costs caused by wireless communications are greater than it is on mobile devices. In this paper, we focus on the problem of energy wastes when tasks are assigned to remote cloud servers or heterogeneous core processors. Our solution aims to minimize the total energy cost of the mobile heterogeneous embedded systems by using an optimal task assignment to heterogeneous cores and mobile clouds. The propose model is named as Energy-Aware Heterogeneous Resource Management Model (EA-HRM2), which is supported by a main algorithm Optimal Heterogeneous Task Assignment (OHTA) algorithm. Our experimental evaluations have proved our approach is effective to save energy when deploying heterogenous embedded systems in mobile cloud systems.