Robust Data Center Network Design using Space Division Multiplexing

With the ever-increasing demand for data transmission in our generation where Internet and cloud concepts play a vital role, it has become essential that we handle data in a most efficient way. A possible solution to overcome the capacity crunch problem which is so evident in future, is applications of Space Division Multiplexing, where we explore the remaining unused domain that is the spatial domain. Space Division Multiplexing using multi-core fibers (MCFs), and few-mode fibers (FMFs) has been studied in our work to enhance the data-carrying capacity of optical fibers while minimizing the transmission cost per bit. The objective of our work is to develop a path protection scheme to handle communication requests in data center (DC) networks using elastic optical networking and space division multiplexing (SDM). Our approach to this problem is to 1) determine a dedicated primary and backup path, 2) possible allocation of spectrum using the flex-grid fixed-SDM model, 3) choose the best possible modulation format to minimize the number of subcarriers needed for data transfer, 4) measure the cost of the resources required to handle the new requests. We propose to evaluate the developed Integer Linear Programming (ILP) formulation based on this scheme, considering the possibility of disasters. We study the impact of the design on the cost of the solution, hence explore whether it promotes significant resource savings.

Name thesis:degree_name

M.Sc.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Windsor

Year dc:date.issued

2018

Author dc:creator

• Biswas, Ankita

Advisors dc:contributor.advisor

• Bandyopadhyay, Subir
• Jaekel, Arunita

Contributors dc:contributor

• djohnst@uwindsor.ca

Licence

• http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language.iso

en_CA

Handle dc:identifier.uri

https://hdl.handle.net/20.500.14776/7605

OAI identifier oai:identifier

oai:uwindsor.scholaris.ca:20.500.14776/7605