Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

An SDN Assisted Framework for Mobile Ad-hoc Clouds release_luhhhjjrqbcnjkyb6btaiecyse

by Venkatraman Balasubramanian, Université D'Ottawa / University Of Ottawa, Université D'Ottawa / University Of Ottawa

Published by Université d'Ottawa / University of Ottawa.

2017  

Abstract

Over a period of time, it has been studied that a mobile "edge-cloud" formed by hand-held devices could be a productive resource entity for providing a service in the mobile cloud landscape. The ease of access to a pool of devices is much more arbitrary and based purely on the needs of the user. This pool can act as a provider of an infrastructure for various services that can be processed with volunteer node participation, where the node in the vicinity is itself a service provider. This representation of cloud formation to engender a constellation of devices in turn providing a service is the basis for the concept of Mobile Ad-hoc Cloud Computing. In this thesis, an architecture is designed for providing an Infrastructure as a service in Mobile Ad-hoc Cloud Computing. The performance evaluation reveals the gain in execution time while offloading to the mobile ad-hoc cloud. Further, this novel architecture enables discovering a dedicated pool of volunteer devices for computation. An optimized task scheduling algorithm is proposed that provides a coordinated resource allocation. However, failure to maintain the service between heterogeneous networks shows the inability of the present day networks to adapt to frequent changes in a network. Thus, owing to the heavy dependence on the centralized mobile network, the service related issues in a mobile ad-hoc cloud needs to be addressed. As a result, using the principles of Software Defined Networking (SDN), a disruption tolerant Mobile Ad-hoc Cloud framework is proposed. To evaluate this framework a comprehensive case study is provided in this work that shows a round trip time improvement using an SDN controller.
In text/plain format

Archived Files and Locations

application/pdf  3.5 MB
file_w34a4lmvj5bytj5n2cpqucqfxe
ruor.uottawa.ca (web)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article-journal
Stage   published
Date   2017-03-31
Language   en ?
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: 005c0b6b-1518-4803-b371-2b2650fe0bc4
API URL: JSON