Latency and Throughput Advantage of Leaf-Enforced Quality of Service in Software-Deﬁned Networking for Large Traﬃc Flows
Devices connected to the internet and networking applications continue to grow each year. As a result, networking infrastructure also grow to be more complex at a rapid rate. Network traffic also becomes bulkier. The Software-Defined Networking (SDN) Architecture has been regarded by many as a potential solution for issues that challenge traditional networking architecture with regards to Quality of Service (QoS) as it provides dynamic, flexible and scalable control and management for networks. A previous study by the authors explored how layers of QoS affect the performance of the network. Results in that study showed a slight advantage by the SDN architecture over the traditional centralized architecture. In this study, that previous study was extended by exploring how additional network traffic will affect QoS performance between the distributed nature of SDN against the traditional centralized nature. Results show that additional network traffic emphasized the performance advantage of the distributed nature of SDN in terms of throughput. With the additional network traffic, the throughput of SDN increased to 100% better compared to the traditional centralized enforcement of QoS. The results have also shown that the results from the VLC streaming media have shown that there were instances were Core-enforced QoS performed better than Leaf-enforced QoS. However, this has been shown as statistically insignificant through the use of the two-sample student’s t-test. Hence, the distributed nature of SDN still performs at least as good as the centralized nature of the traditional networking architecture with regards to the streaming media.
New item … Regencia, J. E. T., & Yu, W. E. S. (2021). Latency and throughput advantage of leaf-enforced quality of service in software-defined networking for large traffic flows. In K. Arai (Ed.), Intelligent Computing (pp. 606–623). Springer International Publishing. https://doi.org/10.1007/978-3-030-80129-8_43