A Single-Link Failure Recovery Approach Based on Resource Sharing and Performance Prediction in SDN

Abstract

Software-defined networking (SDN) is a great innovation, which makes the network programmable so that it is easier to achieve failures recovery. Through pre-programmed recovery strategies and pre-deployed backup resource, flows can be redirected to destination quickly upon failures. However, it consumes a large amount of backup resources for fast failover. Nowadays the network scale and the magnitude of flows increase greatly, which leads to the need to deploy a large number of flow entries in face of failures. However, the Ternary Content Addressable Memory (TCAM) that stores flow entries is capacity-limited. Therefore, it is meaningful to reduce the backup resource consumption. In this paper, we propose a ring-based single-link failure recovery approach (RSFR) to improve backup resource utilization. We select a ring from the network based on node importance and link performance. Then based on the selected ring, we plan backup path and design flow tables to reuse most backup flow entries. Thus failure recovery can be achieved with less flow entries. Additionally, in order to ensure the performance of backup path, we periodically update the ring according to the predicted load. Simulation results show that the proposed approach has a great performance in terms of resource consumption and backup paths performance.