Seminar: Proactive Fog Computing Framework Using Sphere of Visibility

Samuel Kingsford Owusu
M.Sc. Candidate
Supervisor: Dr. K. Vidyasankar

Proactive Fog Computing Framework Using Sphere of Visibility

Department of Computer Science
Wednesday, April 4, 2018, 11:00 am Room: EN 2022


Abstract
Internet of Things-IoT promises a digital connected world, a world where all “things” are connected. This will allow for efficient monitoring and quick real time decision making necessary to ensure smooth and continuous operation of all “things”. In such a system, quality of service (QoS) must somehow be guaranteed. Processing of large volume of stream data in the cloud as initially proposed in IoT comes with several challenges which does not only makes the system inefficient but also costly among others. Leveraging Fog Computing for IoT, these challenges can be addressed. Fog Computing, a service-oriented paradigm, ensures 1) Distributed real time low latency computations, 2) Continuous operation in the absence of network connectivity between sensors and the cloud, 3) Reducing the cloud’s workload by streamlining data before transferring to the cloud. This also helps reduce the bandwidth required to transfer data. Owing to the limited resources of fog computing, task distribution and computation must be “smartly” or “intelligently” carried out in the fog nodes.

In this paper, we propose a conceptual generic application module framework for reducing latency in fog computing through smart distribution and computation of task. The aim of this framework is to fully utilize fog computing resources by doing as many computations as possible in the fog computing environment before data is being sent to the cloud if necessary. Our framework is based on the Concept of Sphere of Visibility. We propose that, effective collaboration among the various fog nodes is required to acquire the desired latency.

Contact

Department of Computer Science

230 Elizabeth Ave, St. John's, NL, CANADA, A1B 3X9

Postal Address: P.O. Box 4200, St. John's, NL, CANADA, A1C 5S7

Tel: (709) 864-8000