Affiliation:
1. Wuhan University of Technology, China & Sokoine University of Agriculture, Morogoro, Tanzania
2. Wuhan University of Technology, Wuhan, China
3. Sokoine University of Agriculture, Morogoro, Tanzania
Abstract
The overwhelming growth of resource-intensive and latency-sensitive applications trigger challenges in legacy systems of mobile cloud computing (MCC) architecture. Such challenges include congestion in the backhaul link, high latency, inefficient bandwidth usage, insufficient performance, and quality of service (QoS) metrics. The objective of this study was to find out the cost-efficient design that maximizes resource utilization at the edge of the mobile network which in return minimizes the task processing costs. Thus, this study proposes a cooperative mobile edge computing (coopMEC) to address the aforementioned challenges in MCC architecture. Also, in the proposed approach, resource-intensive jobs can be unloaded from users' equipment to MEC layer which is potential for enhancing performance in resource-constrained mobile devices. The simulation results demonstrate the potential gain from the proposed approach in terms of reducing response delay and resource consumption. This, in turn, improves performance, QoS, and guarantees cost-effectiveness in meeting users' demands.
Subject
Computer Networks and Communications
Reference50 articles.
1. Borcoci, E. (2016). Fog Computing, Mobile Edge Computing, and Cloudlets - which one? SoftNet Conference, 1-122.
2. Energy-aware fog and cloud interplay supported by wide area software defined networking
3. On the Integration of Cloud Computing and Internet of Things
4. A Stochastic Model to Investigate Data Center Performance and QoS in IaaS Cloud Computing Systems
5. Carlini, S. (2016). The drivers and benefits of edge computing (White paper 226). Schneider Electric –Data Center Science Center. Retrieved September, 22, 2018, from https://www.apc.com/salestools/VAVR-A4M867/VAVR-A4M867_R0_EN.pdf?sdirect=true
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献