Using Osmotic Services Composition for Dynamic Load Balancing of Smart City Applications

Souza, A; Wen, Z; Cacho, N; Romanovsky, A; James, P; Ranjan, R

doi:10.1109/SOCA.2018.00029

Using Osmotic Services Composition for Dynamic Load Balancing of Smart City Applications

Lookup NU author(s): Dr Zhenyu Wen, Emeritus Professor Alexander Romanovsky ORCiD, Professor Raj Ranjan

Downloads

Accepted version [.pdf]

Licence

This is the authors' accepted manuscript of a conference proceedings (inc. abstract) that has been published in its final definitive form by IEEE , 2018.

For re-use rights please refer to the publisher's terms and conditions.

Abstract

Edge computing takes computation away from the Cloud closer to the physical world. Therefore, it reduces thecost of communication bandwidth between IoT devices and the Cloud. However, Edge computing imposes certain limitations in computation power because due to poor hardware capacity of the devices. This restriction may significantly affect the performance of the deployed applications, especially Smart City applications. This limitations also could be aggravated by unpredictable human behaviors which will easily make the Edge computation node overloaded. Osmotic computing is a new IoT application programming paradigm that provides an opportunity to balance the workload between Edge and Cloud therefore to overcome the load imbalance problem of Smart City applications. To this end, we propose an Osmotic Execution Framework that leverages state-of-the-art microservices techniques to deploy and executea Smart City application in a distributed environment including Edge and Cloud. Finally, we evaluate load balancing through latency time analysis of our framework with a real-world smart parking application.