TRAP@NCI

Elastic SDN Based Switch Migration Algorithm to Dynamically Update Map Tables with Minimum Latency and Achieve High Throughput

Balakrishna Deshpande, Chaitanya (2018) Elastic SDN Based Switch Migration Algorithm to Dynamically Update Map Tables with Minimum Latency and Achieve High Throughput. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (2MB) | Preview

Abstract

Software Defined Networks is proving to be the future of networking world and is here to stay. The traditional networks are getting replaced by SDN which mainly improves the exibility of routing the packets. SDN emerged with an idea of having a separate network control plane from its forwarding plane. The control plane is the one which controls the entire network with its centralized controller residing in it. Its main task is to monitor the network behaviour and implement the network policy. Coming to the forwarding plane, i.e, the data plane, it just forwards the packets as indicated by the controller. This paper is focused on building a switch migration algorithm which migrates the switches to the controllers which are not fully loaded and thus having a balanced traffic in the network when the traffic is at its peak. Having a single centralized SDN controller increases the problems with scalability as it takes more time to respond back to the switch. This problem gave way to have a logically distributed multiple controllers which allowed switches to have a better communication and response time with the controllers. But this approach faces problems with static mapping between the switch and a controller which is unable to adapt to traffic variations. Hence multiple controllers at times, are over loaded and eventually causes delay resulting in less throughput. Therefore, to solve all the above problems, this paper showcases the use of Elastic SDN controller which dynamically scales the controller pool depending on the network behaviour. With this approach, latency issue is minimized with much better improvement in the throughput giving 80% efficiency after migrating the switches.

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 NI MHAICIN
Date Deposited: 24 Oct 2018 08:44
Last Modified: 24 Oct 2018 08:44
URI: http://trap.ncirl.ie/id/eprint/3302

Actions (login required)

View Item View Item