论文部分内容阅读
基于流表的转发机制为OpenFlow提供了灵活的可编程能力,但是随着网络功能的不断膨胀,OpenFlow交换机中的流表规模呈现出不断增长的趋势,这些流表难以在交换机有限的三态内容寻址存储器(TCAM)中进行存储,成为网络发展的一个瓶颈。为了高效地利用有限的TCAM资源进行流表存储,该文提出一种OpenFlow多级流表结构及其映射算法,将单一流表映射到多级流表中进行高效存储和查找。仿真结果表明:该文所提方法比单一流表的存储方法节省17%~95%的TCAM资源。这对OpenFlow数据平面查找结构及其扩展性设计具有重要意义。
Flow-based forwarding mechanism provides OpenFlow with flexible programmability. However, with the continuous expansion of network functions, the scale of flow tables in OpenFlow switches shows an increasing trend. These flow tables are difficult to implement in the limited tri-state content of the switch Addressing memory (TCAM) for storage, a bottleneck in the development of the network. In order to efficiently use limited TCAM resources for flow table storage, an OpenFlow multi-level flow table structure and its mapping algorithm are proposed in this paper. A single flow table is mapped into a multi-level flow table for efficient storage and search. The simulation results show that the proposed method saves 17% -95% of the TCAM resources than the single flow table storage method. This is of great importance to the OpenFlow data plane lookup architecture and its extensibility design.