Research On Traffic Engineering Based On SDN

With the rapid development of the Internet which exceed the expectation of people in advance, the comprehensive popularization of social networking, the Internet of things and mobile Internet, and the rise of cloud computing and big data applications, the network traffic has been growing at high speed nowadays. On one hand, the rapid growth of traffic leads to more energy consumption, which would bring unfavorable influence on the environment, economy and sustainable development of the world today. On the other hand, the increasing growth of traffic makes the existing network bandwidth resource more and more intense, as a result, the network would have a lot of congestion frequently. Traffic engineering is an important mechanism to optimize the performance of a data network by dynamically analyzing, predicting, and regualting the behavior of the transmitted data. As using traffic engineering, we can make the most of the existing network resource, reduce the energy consumption of the Internet and improve the overall performance of the network.However, the structure of the traditional networks become increasingly complex and difficult to manage. The forwarding mechanism of the network nowadays is distributed which cannot achieves a fine-grained way of control. And it is difficult to improve the flexibility and scalability of the network. There are a lot of shortcomings of the existing traffic engineering research as well. For example, the link weight optimization cannot make full use of network resources, and MPLS leads to much more overhead of the management load. Therefore, we need a new network architecture and more intelligent and efficient traffic engineering technology. Software Defined Network(SDN) is an emerging networking paradigm that separates the network control plane from the data forwarding plane with the promise to dramatically improve network resource utilization, simplify network management, reduce operating cost, and promote innovation and evolution. The advantage of logical centralization of control layer in SDN makes us have a whole view of the entire network, which can regulate the traffic with high granular and increase the performance of traffic optimization.This paper mainly studies two aspects of network traffic engineering based on SDN, including the energy efficiency and load balancing. First of all, we studies the full SDN traffic engineering problem from the direction of energy-saving. By analyzing the network energy consumption model and using forwarding route through the control of SDN controller, we can realize arbitrary proportion of shunt. Through establishing the mathematical optimization model of the full SDN energy-saving traffic engineering, we propose an energy-saving algorithm. To achieve the goal of energy saving, we close appropriate links and use as less links as possible to concentrate the traffic。Since we need to modify the existing network thoroughly to installed SDN routers all in the network, it is not realistic to achieve this goal in a short period of time. It should be a gradual process to introduce SDN to the traditional IP network. In a hybrid SDN/IP network, both SDN routers and IP routers will exist. This paper studies the energy efficiency and load balancing traffic engineering in hybrid SDN/IP network scenarios. According to the characteristics of SDN and IP network, we guide the network traffic and combine the flow distribution of SDN routers and link weight optimization of IP network. At the same time, we establish optimization model from the aspects of energy-saving and load balancing, and propose corresponding algorithm, which can get better effect of energy efficiency and load balancing, respectively.