Detection and Re-routing of elephant flows in a Software Defined Networking to avoid traffic congestion

Bavugi, Sainath (2018) Detection and Re-routing of elephant flows in a Software Defined Networking to avoid traffic congestion. Masters thesis, Dublin, National College of Ireland.

Preview

PDF (Master of Science) Download (1MB) | Preview

Abstract

Software Defined Networking (SDN) is widely used in datacenters because it makes the network topology exible. It is very crucial to monitor and manage both incoming and outgoing traffic in order to utilize the resources that are available and avoid congestion. In a network, a large/elephant flow absorb more network bandwidth as well as takes time to get processed which impacts QoS. In this paper, a variety of approaches are discussed for detection of elephant flows and routing algorithms. The proposed routing algorithm scans the network to find all the paths available between source and destination and calculates the link bandwidth of different paths available. This research paper used mininet as network emulator, sFlow for traffic analysis and POX controller along with the OpenFlow protocol to identify the network paths. The algorithm has been tested on a fat tree topology and evaluated against different scenarios. In all the test scenarios, it was possible to detect elephant ows and the algorithm finds first and second best path out of available paths. Ultimately, the second best routing path is displayed which could be opted for the large flow.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Caoimhe Ní Mhaicín
Date Deposited:	24 Oct 2018 09:00
Last Modified:	24 Oct 2018 09:00
URI:	https://norma.ncirl.ie/id/eprint/3303

Actions (login required)

View Item