Study on QoS-enabled Adaptive Transmission in Software Defined Networks

• Main Authors: Yu-Shan Tsai, 蔡育珊
• Other Authors: Yeong-Sheng Chen
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/09486208463708612910

碩士 === 國立臺北教育大學 === 資訊科學系碩士班 === 103 === The software defined networks (SDN) has more flexibility than the legacy networks because it is a programmable network with a controller acting as the brain of a network and orchestrating the traffic in the network by providing forwarding information to the forwarding units such as switches. In general, the traffic through a network is generally stochastic in nature. This is reflected in dynamically varying network parameters like available bandwidth and delay of links. Stochastic switching is an approach in which choosing the egress port for traffic towards a specific destination among possible outgoing ports follows a probability distribution. This probability might change based on the current network behavior. The purpose of this work is to investigate the feasibility of enhancing video transmission in a SDN environment with stochastic switching. Traditionally, the controller provides forwarding information in a deterministic manner. For example, the controller may direct the forwarding device to forward all the packets from ingress port A to a particular outgoing port B in a certain time period. However, instead, this study presents an optimization scheme for the controller to determine forwarding decisions in a probabilistic manner so as to enable dynamic Quality of Service (QoS) over SDN networks. Simulation results verified that the proposed scheme achieves significant improvement in overall quality of data transmission under various traffics with different QoS levels.