Backup-resource Based Failure Recovery Approach In SDN Data Plane

In the past decade, with the great progress in the field of information and communication technology (ICT), the traditional Internet is facing serious challenges. For example, the rapid development of wireless technology has promoted the field of mobile management continue to launch new research; large data center is facing serious energy consumption problems; time-sensitive cloud business needs the support of high-quality underlying network services. In recent years, in order to solve these problems, Software Defined Networking (SDN), as a key research object, has attracted great interests both in academia and industry. The core idea of SDN is to separate control plane and forwarding plane, and to introduce the idea of network programming at the same time. The SDN controller, as a network operating system, detaches the remote controller from the underlying devices, which allows people to easily program the network anytime and anywhere. The good features of SDN enable new services to be deployed quickly and new academic research methods can be easily implemented. At present, the OpenFlow protocol is the most potential technology to standardize the communication between the controller and the network devices. At the same time, the widespread use of the OpenFlow protocol makes the SDN attract more and more attention from scholars.The rapid development of the SDN brings many challenges, and one important problem is failure recovery. If a link or a node fails, the switches that can detect the failure have to either inform controller to update flow tables or transform the data to pre-configured paths to recover the failure.And existing failure recovery approaches mainly consider the recovery delay and packet loss, and ignore the storage resources consumption for backup paths in case of link or node failure. However, these methods do not consider the following two problems: one is the storage resources consumption for backup paths in case of link or node failure, another is that that the backup paths may bring serious load for network, which may influence performance of the whole network. Based on the above discussion, the main contributions of this thesis are as follows:(1) In order to minimize the consumption of backup resources and meet the required failure recovery delay, a backup-resource based failure recovery approach is proposed. Two metrics are proposed to grade physical links, and three kinds of strategies for different graded links are provided,based on which the approach tries to use less flow entries to recover link failure and meets the required failure recovery delay, while guaranteeing the reliability of the network. Simulations show that backup-resource based approach can use as less flow entries as possible to ensure the performance of failure recovery and satisfy the required delay of important traffic at the same time. Moreover, the approach has good and steady performance in networks of different scales and connectivity.(2) In order to balance the load and meet the required recovery delay,a failure recovery approach with load-balancing is proposed. The traffic flows are classified according to the requirement of recovery delay, and two kinds of strategies with load-balancing for different graded traffic flows are provided, based on which the approach tries to reduce the influence on network performance produced by the load of backup paths and meets the required failure recovery delay. Simulations show that the proposed approach can timely adjust backup paths before failure occurring according to present network load so as to ensure the load-balancing of link resources for failure recovery in SDN network.