MpFPC–A Parallelization Method for Fast Packet Classification

Abstract

Packet classification is the core technology of network layer and an important means to ensure the security of network system. With the rapid development of network technology, higher requirements are put forward for the speed of network packet classification. This paper improves the traditional single thread package classification framework, A new parallelization method for fast packet classification (MpFPC) based on distributed computing is proposed, the method adopts the packet classification idea based on decision tree, but compared with the traditional algorithm, a rule mapping preprocessing process is added before constructing the classification decision tree, which effectively removes the rule redundancy and conflict, so as to avoid the rule replication problem of the traditional decision-tree-based method. In addition, the method can group the rules and data packets at the same time, which improves the packet classification efficiency. Experimental results show that MpFPC method has high classification efficiency and has obvious speed advantage compared with Uscuts method with time complexity of O(klogn). In addition, the test results also show that the classification speed of MpFPC will increase with the increasing number of computing nodes, which provides a new possible way to meet the classification wire-speed requirement.

PcmSU-A Packet Classification Method Supporting High-Speed Search and Fast Update

“…if d='1' then perform steps (21) in the R.eq interval; perform steps (18) in the R.ad interval; 34. if d='0' then perform steps (22) in the R.eq interval; perform steps (19) in the R.ad interval; } //End of PruningTree(R, T , d) and if there is no overlap between the rule interval and the node interval of the decision tree on a certain dimension, then no pruning operation is required in subsequent dimensions. In addition, compared with the traditional decision tree-based classification method, the PcmSU method has another advantage: as the decision tree is divided into several subtrees, its scale must be reduced to a certain extent, which on the one hand reduces the complexity of decision tree reconstruction when rules are updated, and on the other hand, it also provides the foundation for distributed packet classification method [28], and improves the efficiency of packet classification. The execution process of the above rule update can be extended to the general k-dimensional rule case, as described in Algorithm 4.…”