Accelerating packet classification with counting bloom filters for virtual OpenFlow switching
Accelerating packet classification with counting bloom filters for virtual OpenFlow switching
复制标题
DOI：
10.1109/cc.2018.8485474
复制
发表时间：
2018-10
影响因子：
4.1
通讯作者：
Jinyuan Zhao;Zhi-gang Hu;Bing Xiong;Keqin Li
中科院分区：
文献类型：
--
作者：
Jinyuan Zhao;Zhi-gang Hu;Bing Xiong;Keqin Li
网络虚拟化的发展趋势导致虚拟交换机在虚拟化环境中的广泛采用。然而，虚拟交换面临着巨大的性能挑战，特别是在基于OpenFlow的软件定义网络的数据包分类。本文首先深入研究了虚拟OpenFlow交换中的数据包分类问题，指出其性能瓶颈主要是对每个到达数据包进行多次失败掩码探测的流表遍历。在此基础上，提出了一种基于计数布隆过滤器的高效数据包分类算法。特别地，计数布隆过滤器被应用于预测具有很大可能性的流表查找的失败，并且对于失败的掩码探测绕过流表遍历。最后，我们提出的数据包分类算法进行了评估与真实的网络流量跟踪实验。实验结果表明，该算法在平均搜索长度方面优于经典算法，有助于提高虚拟OpenFlow交换性能。
The growing trend of network virtualization results in a widespread adoption of virtual switches in virtualized environments. However, virtual switching is confronted with great performance challenges regarding packet classification especially in OpenFlow-based software defined networks. This paper first takes an insight into packet classification in virtual OpenFlow switching, and points out that its performance bottleneck is dominated by flow table traversals of multiple failed mask probing for each arrived packet. Then we are motivated to propose an efficient packet classification algorithm based on counting bloom filters. In particular, counting bloom filters are applied to predict the failures of flow table lookups with great possibilities, and bypass flow table traversals for failed mask probing. Finally, our proposed packet classification algorithm is evaluated with real network traffic traces by experiments. The experimental results indicate that our proposed algorithm outperforms the classical one in Open vSwitch in terms of average search length, and contributes to promote virtual OpenFlow switching performance.