2013-09 MSc Dissertation

Abstract:

Throughout this research project, the efficiency, reliability and practicality of popular networking protocols are examined and analysed. Given the age of the most common protocols used on networks and the Internet today; are they still suitable considering the now unprecedented scale of the Internet, can they support the increasing demands of modern day networks for millisecond convergence and fault recovery, do they efficiently transport bandwidth orders of magnitude larger than ever believed possible? These are the kind of questions this research project looks at, studying the original designs and use purposes of Ethernet, IPv4 and TCP (amongst others), comparing them to their modern superseding revisions and increased operational complexity. It also looks at the network services they carry and the demands of present day networks, investigating where the protocols fall short of their requirements, how we can detect this, and how these scenarios can be rectified or mitigated. This is achieved by first compiling a baseline from protocol standards, guidelines and deployment statistics. Research on known existing protocol issues is brought under investigation to assist in looking for protocol design under sight and efficiency limits. The open views of the networking industry are then added in for a more qualitative insight. Experimental research builds upon these ideas to ratify them and create conclusive data. Conclusions are drawn from the research findings which provide a clearer picture on the current operational status of the key networking protocols and their optimal deployment scenarios.

PDF of final dissertation

---