基于联合多维动态Bloom过滤器的SDN多属性查找结构研究

软件定义网络(SDN)作为一种新型的网络架构,可以为核心网络及应用的创新提供良好的平台.作为SDN实现的关键技术Openflow在实际的应用中仍然存在一些技术瓶颈.本文提出一种联合多维动态Bloom过滤器(CMDDBF)查找结构来替代Openflow交换机中的三态内容寻址存储器(TCAM)来实现流表的转发和处理.CMDDBF解决了TCAM空间利用率低和价格昂贵的缺点,能够有效实现多字段数据的查找和处理操作,并根据流表的规模动态调整内存空间.实验表明:CMDDBF不仅可以有效实现多字段数据的查找,而且在降低误识率方面具有很好的性能.

Research on SDN multi-attribute query structure based on combined multi-dimension dynamic Bloom filter

As a new innovation structure of network,software-defined networking (SDN) provides a good platform for the core network and application innovation.However,Openflow as a key of technology to implement SDN still exists some technical bottlenecks in the actual application.In this paper,we proposed using Combined Multi-Dimension Dynamic Bloom Filter(CMDDBF) to replace the TCAM(Ternary Content Addressable Memory) in Openflow switch to forward and process the data of flow tables,which can solve the shortcomings of TCAM such as lower space utilization and expensive.In addition,CMDDBF can dynamic adjust memory space according to the size of data and it is more suitable for query and process operations of flow table.Experimental analyses show that CMDDBF can not only realize the query of multi-attribute data effectively,but also have a lower false positive.