Path Establishment in Software Defined Optical Networks

Publication Type:
Thesis
Issue Date:
2023
Full metadata record
A newly emerged networking paradigm called Software Defined Networks (SDN) is an architecture to overcome the problems with conventional data networks. SDN renders a solid potential for flexible network control and management. In particular, the standard OpenFlow protocol is often referred to as a primary new notion in the networking domain. SDN presents the idea of separating the control plane decisions from the data forwarding plane and gives the concept of programmable networks by forming a standard programming interface. A centralized controller makes routing decisions, while the switches only perform forwarding actions. It allows testing new protocols and algorithms to control data packets. The SDN technology provides ease of control of the network’s infrastructure, is cost-efficient and open and enables programmable components. SDN is very efficient at providing flexible control in packet networks. However, in optical networks, circuit-switched paths are planned and set up much before the arrival of any data and any change of route or service providers will require a long, slow process and manual intervention. Previous solutions (GMPLS/ASON etc.) to automate optical networks have become very complex. Adopting SDN for optical networks requires initiating an intermediate system to manifest the network as an SDN-capable domain. This research aims at developing an intermediate system that virtualizes layer one optical network to the SDN controller as a network of layer two OpenFlow switches. It allows the programmability of paths through optical switches in an on-demand fashion using OpenFlow protocol and eliminates the need for traditional remote and local decision-making algorithms. This contributes to the dynamic service provision in optical networks. Our model offers a translation mechanism between controller and optical switches so that the controller can control path computation without being informed of underlying structure and wavelengths. Adding programmability to the network environment will bring scalability, adaptability and robustness. The translation mechanism is integrated into the middle of the SDN architecture to provide an abstraction layer to manage the flow setups, allowing them to adapt dynamically according to real-time demands and needs. The component-based model provides room for adding more features in the future. This dissertation is presented in three parts. In part 1 motivation for the research and the problem background are presented. The proposition is proved in the second part of the thesis (chapter 4) by implementing the mode. Finally, the project is concluded in part 3 of this dissertation, and future directions are provided (chapter5).
Please use this identifier to cite or link to this item: