| Summary: | 碩士 === 國立交通大學 === 網路工程研究所 === 106 === Software-defined network (SDN) is a new network paradigm that allows traffic engineering and management by decoupling the control and data planes. A network operator can flexibly program packet forwarding behaviors, which can be translated into rules and actions installed in configurable switches. Therefore, SDN-based traffic engineering and flow management have attracted a lot of attention recently because of its centralized control, flow programmability and flexible resource management. However, most of the advanced designs require per-flow management. When a network scales up and includes a large number of flows, the number of forwarding rules in a switch could exceed the limited flow table space and also lead to undesired processing overhead. To overcome such huge traffic demands, in this thesis, we propose a cluster-based customized flow management framework to balance the trade-off between customized services and management costs. The key idea of our design is to group similar flows into a cluster and replace per-flow management with per-cluster management. To achieve this goal, we cluster flows with consideration of the similarity of their traffic patterns. In our cluster-based design, the number of flow rules now becomes proportional to the number of clusters, instead of the number of flows. Hence, to adapt to varying traffic load, we further propose a hierarchical cluster merging scheme to dynamically merge clusters and make the best tradeoff between flow entry saving and routing performance guarantee. We evaluate the performance of our designs via trace-driven simulations. The results demonstrate that the proposed cluster-based flow management framework can significantly reduce the flow table usage, while producing minimum negative impact on routing performance.
|
|---|
