Fast Failure Recovery for In-Band Controlled Multi-Controller OpenFlow Networks

Abstract

Benefit from central control paradigm, SDN owns flexible and agile controllability to the networks. However, the central controller becomes the most vulnerable device. The failure of the controller results in the malfunction of the whole network. OpenFlow specification allows a network to equip multiple controllers to enhance the reliability on the control plane. How to perform fast failure recovery that takes advantage of a network with multiple controllers especially for networks operating with in-band control has not been well investigated. In this paper, we design a fast failure recovery scheme that takes the multiple controller architecture into consideration. In our system, there is a main controller that is responsible for controlling the network in the normal state. The other controllers are standby controllers used to take over the network control in failure state. The failure recovery in the proposed scheme includes three phases- fast failure detection, fast failure location identification, and traffic reroute. The fast failure detection uses monitoring cycles to examine the status of the network. As a failure is detected, the main controller and the standby controllers work together to perform failure location identification to pin-point the locations of the failure devices. In the traffic reroute phase, the main controller and the standby controllers collaborate to complete the fast failover. A control path provisioning algorithm is proposed to resolve the disjoint path planning problem for the routing of the in-band control channels from the controllers to the switches. As a result, a failure can be fast recovered even if it damages both working control paths and data paths. We have implemented our design in an experimental network. The experimental results show that the proposed scheme can protect any single link or single node failure within 50 msec.