A Simulated Annealing-based Failover Mechanism for Hierarchical SDN Controllers

• Main Authors: Hsieh, Hsiao-Hu, 謝曉湖
• Other Authors: Wang, Kuochen
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/h3b77q

碩士 === 國立交通大學 === 網路工程研究所 === 105 === The multi-controller architecture is a must to improve scalability and reliability in software defined networks (SDNs). A hierarchical multi-controller architecture allows a global controller to handle rare events, such as routing events from local controllers. A failover mechanism, which includes failure detection and failure recovery, is necessary for the multi-controller architecture in case of a controller failure. Existing controller failure detection and failure recovery mechanisms are performed in local controllers, which may degrade the performance of the local controllers. In failure recovery, existing controller failure recovery mechanisms consider two metrics with fixed weights, switch-controller propagation delay and load standard deviation among controllers, to redo switch-controller association. However, one needs to pay more attention to the time-varying load standard deviation metric in order to reduce flow setup time. To address these problems, we propose a simulated annealing-based failover mechanism (SAFM), run as an additional event in global controllers, to relieve the burden of local controllers for high availability hierarchical SDN controllers. SAFM uses multiple global controllers to detect a local controller failure and compute a switch migration plan to redo switch-controller association for controller failure recovery. In failure detection, global controllers exchange their failure detection results of local controllers to jointly make a local controller failure decision to reduce mistake rates. In failure recovery, the proposed SAFM considers two metrics: switch-controller propagation delay and load standard deviation among local controllers with adaptive weights to adapt to time-varying local controllers’ loads. In addition, in our design, a switch is connected to two local controllers so that asynchronous messages, such as packet-in and port-status messages, will not be lost during controller failover. Experiment results show that, SAFM has the best load balance result in term of load balance metric (LBM) of 1.035, while the best LBM of related works is 1.087. SAFM also has the lowest flow setup time of 7.837 ms, while the best of related works is 10.418 ms.