Home // ICNS 2019, The Fifteenth International Conference on Networking and Services // View article

Least Loaded Sharing in Fog Computing Cluster

Authors:
Sammy Chan

Keywords: fog computing; load balancing; collaboration servers; buffer sharing

Abstract:
Recently, there has been a growing ubiquity of connected devices and sensors in wireless sensor networks, health care systems, smart grids and smart cities, forming the Internet of Things (IoT). IoT devices generally have limited computation resources, and thus rely on the computational and storage resources of the cloud. However, IoT applications generally have a real-time requirement that cannot be fulfilled by mainstream cloud services. Therefore, a new paradigm called {em fog computing} has emerged to offload the computation and storage needs of end user devices to the servers in the network edge. In this paper, we propose a least loaded sharing method to fully exploit the collaboration between fog servers and to achieve load balance among them. In our method, an overloaded server is able to react to temporary peaks of requests by forwarding the incoming requests to the least loaded neighbour server. The proposed method helps to reduce the blocking probability of requests and the delay experienced by accepted requests. We also develop a computationally efficient analytical model to evaluate the performance of our proposed method.

Pages: 27 to 31

Copyright: Copyright (c) IARIA, 2019

Publication date: June 2, 2019

Published in: conference

ISSN: 2308-4006

ISBN: 978-1-61208-711-5

Location: Athens, Greece

Dates: from June 2, 2019 to June 6, 2019