TEAM: Time and Energy Aware Application Partitioning and Resource Allocation Strategy on MEC Platform

In recent times, the hand-held smart devices are entrusted with running several data and compute-intensive applications which are often of critical nature. As a result, the mobile edge computing (MEC) paradigm becomes popular that facilitates an efficient execution mechanism for these applications. The existing scheduling policies in this execution platform mainly considers independent user tasks and supports only full offloading. In this paper, we propose a scheduling strategy for a bag of applications supporting partial offloading to reduce the energy consumption of the edge hosts. Further, for the execution of applications in edge host, we propose an execution model using light-weight container-based solution Kubernetes in place of traditional virtual machine-based model. Experimental results show the effectiveness of our proposed scheduling policy with an energy saving of 7.9% with respect to the baseline state-of-art energy-efficient scheduling policy. The proposed policy also meets the delay constraints of all the applications.