NORMA eResearch @NCI Library

Performance Optimization of Software Defined Networks By Dynamic Placement of Controllers

Syed, Imran Abdul Rahman (2020) Performance Optimization of Software Defined Networks By Dynamic Placement of Controllers. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (4MB) | Preview
[img]
Preview
PDF (Configuration manual)
Download (127kB) | Preview

Abstract

Software Defined Networking is a new model in the era of cloud computing. It provides programmability which in turn helps to configure and administer a network with continuously changing network state. Software Defined Networking has logical plane separated from the data plane. The logical plane is centralized where all the decision making happens. The data plane receives logical decisions from the control plane and works on forwarding of data packets. The control plane consists of the controllers and the data plane consists of the switches. A typical SDN has a controller which handles the routing decisions and switches that work on the decisions made by the controller. However, this has constraints as a single controller can be overloaded and become a single point of failure. This reduces scalability and performance of the whole network. Recent works have been put forward to mark these issues. Most of them suggested the placement of multiple controllers, instead of a single one, which can work concurrently to serve the switches better. However, this approach raises another problem. Since the network’s state keeps changing, placing of multiple controllers becomes more challenging. Hence, the problem of Dynamic Controller Placement comes in to picture. The number of controllers in a network should be increased dynamically depending on the number of switches and hosts. This will lead to load balancing among the controllers. When the controllers in a network are balanced, the performance increase can been seen in terms of increased throughput and reduced latency.
Keywords: Software Defined Networking, Dynamic controller placement, Controller Load Balance, Throughput

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions: School of Computing > Master of Science in Cloud Computing
Depositing User: Dan English
Date Deposited: 29 Jan 2021 12:16
Last Modified: 29 Jan 2021 12:16
URI: http://norma.ncirl.ie/id/eprint/4556

Actions (login required)

View Item View Item