A novel offloading framework for computation service in the Edge Cloud and the Core Cloud: A case study for face recognition

Abstract

A fast rate of progress has allowed the proliferation of smartphones and ease their extensive presence in people’s daily life. However, inadequate processing speed, and limited battery capacity have hindered improvements of the computational capabilities of the smartphone. Offloading computational tasks to the remote Cloud (Edge and Core) could solve this problem by enabling the user to access these services over the Internet. Edge Cloud computing has been recognized as an emerging field within the Cloud computing paradigm, where computation servers are situated at the edge of the Internet to reduce network delay and traffic. Nevertheless, offloading tasks to the remote Cloud is not always beneficial due to variable network conditions and added processing costs.

In this thesis, a framework is proposed that provides smartphones with the ability to make offloading decisions to minimize processing time or cost, energy consumption or any combination of these three parameters. To validate the accuracy of the framework, face recognition is used as an application and is implemented in the remote Cloud infrastructures. Experimental results based on real-life and alternate scenarios demonstrate that the framework can make the correct offloading decision, in a cost effective way.