EAKOS - Energy Aware KMeans Offloading Scheme for Mobile Applications

Recently Mobile Cloud Computing has been considered an emerging trend to handle mobile applications that run on resource constraint mobile devices. Computation offloading is a major issue in mobile Cloud Computing where the performance can be enhanced by reducing the energy consumption of mobile applications. In this context, several solutions have been proposed to optimize computation offloading and the parameters to be considered while performing the operations. Identifying the modules of the mobile application for offloading to Cloud Infrastructure under different parametric conditions remains a challenge. Thus, this paper proposes Energy Aware KMeans Offloading Scheme (EAKOS) a dynamic computation offloading scheme with optimal partitioning using KMeans Clustering based on the frequency of use of the modules by the user, computation resource requirements, and the remaining battery power on the Mobile Device. The results show that this scheme can be beneficial for enhancing the performance of mobile applications with a 69% reduction in battery power consumption than the existing offloading schemes.