Enhanced bandwidth management model using software defined networking

Emilia Rosa Jimson, (2019) Enhanced bandwidth management model using software defined networking. Masters thesis, Universiti Malaysia Sabah.

Enhanced bandwidth management model using software defined networking.pdf

Download (962kB) | Preview


Available bandwidth is the maximum unused bandwidth at a link, whereby it is measured in bits per second. However, since available bandwidth in network is limited, it is very critical to manage the network bandwidth efficiently. In this research, bandwidth management model based on Software-Defined Networking (SDN) architecture is examined. Through the proposed model, real-time traffic was given priority to access the limited bandwidth over non-real-time traffic. SDN promises the simplification of network management compared to the traditional network. SDN architecture breaks the vertically integrated current network design and it introduces centralized network control through the SDN controller, which is considered as the greatest contribution of SDN in networking area. SDN architecture is adopted in this research because it promotes centralized control features which enable network programmability and supports better network resources management. Simulation was conducted using the EstiNet network simulator. Two groups of experiments with different situations have been conducted. Each experiment consisted of five simulations with different number of sender and receiver nodes. The first experiment simulated the traffics in a traditional network; and the second experiment simulated the traffics in a proposed network model. In this research, most of the available bandwidth was reserved to real-time traffic and the remaining portion was reserved for non-real-time traffic. The real-time throughput results of the two different experiments have been compared to evaluate the performance of the proposed model. The experiment results showed that the proposed model has successfully given priority to real-time traffic to access the limited network bandwidth. As the number of nodes increases, the average real-time throughput of experiment one and experiment two decreases linearly. When nodes reaches 10, the real-time average throughput of experiment one and experiment two are 635KBps and 473KBps respectively.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Network , Software-Defined Networking (SDN) , architecture , programming
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: FACULTY > Faculty of Computing and Informatics
Depositing User: Noraini
Date Deposited: 11 Mar 2020 06:05
Last Modified: 11 Mar 2020 06:05
URI: http://eprints.ums.edu.my/id/eprint/25139

Actions (login required)

View Item View Item