Exploring the Effects of Network Topology Layers on Quality of Service Mechanisms in the Context of Software-Defined Networking

• Main Authors: Regencia, Josiah Eleazar T, Yu, William Emmanuel S
• Format: text
• Published: Archīum Ateneo 2020
• Subjects: Software-Defined Networking; Quality of Service; Class- based Queueing; Computer Sciences; Databases and Information Systems
• Online Access: https://archium.ateneo.edu/discs-faculty-pubs/205; https://doi.org/10.1007/978-981-33-6385-4_2
• Institution: Ateneo De Manila University

The separation of the control plane and the data plane in the Software-Defined Networking (SDN) architecture makes it easier and more feasible to implement per-flow Quality of Service (QoS) provision- ing in the network which is an advantage when dealing with multimedia flows such as VoIP and video streaming, etc since these require special QoS handling. Previous studies have explored applying QoS mechanism to a distributed single-layered set of switches in the context of SDN have shown promising results. In this study, the focus is to explore how layers in the network topology affect the performance of QoS in the network. The purpose for such is in order to replicate real-world networking sce- narios. A test framework for QoS algorithms and networking topologies in the context of SDN that was introduced in a previous study was used to perform the experiments. Generally, results have shown that having less layers in the network means better performance. However, this does not mean that adding layers create a material disadvantage. Results also show that despite adding layers to the networking topology, the SDN dis- tributed Leaf-enforcement of QoS performed better or as good as com- pared to the traditional centralized Core-enforcement of QoS. It is noted that there are instances of Core-enforced tests that performed better than Leaf-enforced tests. However, the student t-tests show that the dif- ference in results is not material hence, there is no sufficient evidence to make such conclusion.