Xiang Wang
ABSTRACT
Most recent works demonstrate that grouping methodology could bring significant reduction of memory usage to decision-tree packet classification algorithms, with insignificant impact on throughput. However, these grouping techniques can hardly eliminate rule-replication completely. This work proposes a novel rule grouping algorithm without any replication. At each space decomposition step, all rules projecting on the split dimension form the maximum number of non-overlapped ranges, which guarantees the modest memory usage and grouping speed. Evaluation shows that the proposed algorithm achieves comparable memory size with less pre-processing time.
- J. Pettit, J. Gross, B. Pfaff, M. Casado and S. Crosby. "Virtual Switching in an Era of Advanced Edges", in Proc. of DC CAVES, 2010.Google Scholar
- B. Vamanan, G. Voskuilen and T. Vijaykumar. "EffiCuts: Optimizing Packet Classification for Memory and Throughput", in Proc. of SIGCOMM, 2010. Google ScholarDigital Library
- J. Fong, X. Wang, Y. Qi, J. Li and W. Jiang. "ParaSplit: A Scalable Architecture on FPGA for Terabit Packet Classification," in Proc. of HOTI, 2012. Google ScholarDigital Library
- Y. Qi, L. Xu, B. Yang, Y. Xue and J. Li. "Packet Clas-sification Algorithms: From Theory to Practice", in Proc. of INFOCOM, 2009.Google Scholar
Index Terms
- Replication free rule grouping for packet classification
Recommendations
EffiCuts: optimizing packet classification for memory and throughput
SIGCOMM '10Packet Classification is a key functionality provided by modern routers. Previous decision-tree algorithms, HiCuts and HyperCuts, cut the multi-dimensional rule space to separate a classifier's rules. Despite their optimizations, the algorithms incur ...
EffiCuts: optimizing packet classification for memory and throughput
SIGCOMM '10: Proceedings of the ACM SIGCOMM 2010 conferencePacket Classification is a key functionality provided by modern routers. Previous decision-tree algorithms, HiCuts and HyperCuts, cut the multi-dimensional rule space to separate a classifier's rules. Despite their optimizations, the algorithms incur ...
Replication free rule grouping for packet classification
Most recent works demonstrate that grouping methodology could bring significant reduction of memory usage to decision-tree packet classification algorithms, with insignificant impact on throughput. However, these grouping techniques can hardly eliminate ...
Comments