论文部分内容阅读
目前Open Flow流表的存储与查找还面临一些挑战,如流表项匹配字段数量的不断增加、流表规模的爆炸性增长以及实现Open Flow硬件交换机的端口线速处理更加困难.针对这些问题,本文提出了一种高效的Open Flow流表存储与查找实现方法.该方法将Open Flow流表划分成多个规模更小、匹配字段更少的子流表,进一步根据字段匹配类型将子流表切分成多个字段,不同字段采用不同的算法和硬件进行存储和查找,得到字段匹配结果后再结合汇总表查找从而得到子流表的匹配结果.通过划分流表与切分子流表来压缩流表存储空间,根据子流表的查找选择和子流表内部查找优化来减少查找过程中硬件资源开销.通过大量模拟实验表明,该存储与查找方法能够压缩50%左右的存储空间,且可以有效减少流表项匹配过程中80%硬件资源的消耗.
At present, Open Flow flow table storage and search also face some challenges, such as the increasing number of flow entry matching fields, the explosive growth of flow scale and the realization of Open Flow hardware switch port wire-speed processing more difficult.In view of these problems, this article This paper proposes an efficient implementation method of Open Flow flow table storage and search.This method divides the Open Flow flow table into several smaller sub-flow tables with smaller matching fields and further analyzes the sub-flow table cut according to the field matching type Divided into multiple fields, different fields using different algorithms and hardware for storage and search, get the field matching results and then combined with the summary table lookup to get the sub-flow table matching results by dividing the flow table and the sub-flow table to compress the flow table Storage space to reduce the hardware resource overhead during the search process according to the search selection of the sub-flow table and the internal search optimization of the sub-flow table. The simulation results show that this storage and search method can compress about 50% of the storage space and can effectively reduce the flow 80% of hardware resource consumption during table matching.