Probabilistic suffix tree based packet classification on FPGA

The general packet classification problem has received a great deal of attention over the last decade. The ability to classify packets into flows based on their packet headers is important for QoS, security, virtual private networks (VPN) and packet filtering applications. Multi-field packet classification has evolved from traditional fixed 5-tuple matching to flexible matching with arbitrary combination of numerous packet header fields. In this project, we introduce a method for producing more efficient probabilistic suffix tree (PST) classifiers. Our empirical model is general enough to model many disparate problems. We considered the next-generation packet classification problems where more than 5-tuple packet header fields would be classified. When matching multiple fields simultaneously, it is difficult to achieve both high classification rate and modest storage in the worst case. Our classifier can be considered among the most algorithms which have high throughput and efficiency.