Abstract:

Aimed at the problem of the rapid growth of the flow table size and the inefficient utilization of the flow table storage resources, this paper proposes a two-step multiple flow table construction algorithm. First, the match fields in the single flow table are split into multiple flow tables according to the flow classification vectors, so the wildcards in the flow table are eliminated. Second, the orthogonal decomposition is performed for each sub-tables, and the redundant entries are compressed. Experimental results indicate that the method can save more than 60% storage space for a given flow table. Compared with existing works, the flow table compression ratio is improved by 21.4% to 51.5%. FPGA verification shows that this method can be implemented in practical hardware and the pipeline processing speed is guaranteed. The data throughput can achieve 197 Million Packets Per Second (MPPS), which supports the 100 Gbit/s line-rate packet processing.

Key words: software-defined network, flow table, storage space optimization, flow classification, field programmable gate array