NFA Reduction for Regular Expressions Matching Using FPGA

FPGA, NFA, Reduction, Regular expressions matching

Many algorithms have been proposed to accelerate regular expression matching via mapping of a nondeterministic finite automaton into a circuit implemented in an FPGA. These algorithms exploit unique features of the FPGA to achieve high throughput. On the other hand the FPGA poses a limit on the number of regular expressions by its limited resources. In this paper, we investigate applicability of NFA reduction techniques - a formal aparatus to reduce the number of states and transitions in NFA prior to its mapping into FPGA. The paper presents several NFA reduction techniques, each with a different reduction power and time complexity. The evaluation utilizes regular expressions from Snort and L7 decoder. The best NFA reduction algorithms achieve more than 66% reduction in the number of states for a Snort ftp module. Such a reduction translates directly into 66% LUT and FF saving in the FPGA.