Collaborative Mobile Edge and Cloud Computing: Tasks Unloading for Improving Users’ Quality of Experience in Resource-Intensive Mobile Applications

The advancement in resource-intensive and latency-sensitive applications challenge the legacy systems in Mobile Cloud Computing (MCC) in terms of network congestion, bandwidth utilization, performance and Quality of Service (QoS) metrics. Such challenges emanate from first, limited energy sources and resource poverty of mobile devices. Second, multi-hop connection between user devices and the cloud. To address such challenges, mobile edge computing is a promising solution. This study proposes an architecture that considers unloading resource-intensive tasks from clients' devices to more resourceful edge servers which exploit cooperative approach for tasks processing. Thus, it is essential for minimizing delay, bandwidth usage, congestion to the core network and guarantees cost-effective approach for meeting user's demands. The simulation results show that the proposed approach through unloading, it reduces response time and energy usage. This in turn improves performance, system utility and Quality of Experience (QoE).

I. Introduction

Resource-intensive mobile applications increasingly become complex worldwide due to emerging trends of data into data-intensive paradigm making legacy systems in Mobile Cloud Computing (MCC) inefficient for managing big data with increased demands of users [1 , 2] . Literature review provide evidence that high bandwidth applications such as video streaming, medical imaging, gaming, social networking, remote access, web browsing and many others has grown rapidly between the year 2014 to 2018 and even more in the coming years [3] . Also, data traffics for mobile applications grow quickly and it is anticipated to grow more in near future [4] . Such big data necessitates intensive resources for computing, storage and bandwidth. Thus, the growing trends pose challenges to MCC architecture in terms of high network congestion, network over-load, bandwidth utilization and performance [5 , 6] . Such challenges in turn, become stumbling block for meeting the demands for low latency, users’ Quality of Experience (QoE), cost-effectiveness, and location awareness.