Abstract
Modern deep packet inspection (DPI) systems match network traffic against a large set of patterns which are defined using regular expressions. Deterministic finite automata (DFA) is generally preferred to parse these regular expressions. However, packets are mostly scanned one byte at a time which becomes a bottleneck for the DPI systems as they are unable to cope up with higher line rates. In this paper, we present an approach which allows a packet to be split up into two chunks. Furthermore, we scan the bytes of each chunk in parallel using speculation and multi-stride (stride-k) DFA. Stride-k DFAs results in fast processing of bytes but leads to high memory usage. Therefore, we propose a transition compression algorithm using alphabet compression table to limit the memory usage of multi-stride DFA. Experimental results show that the speculative parallel pattern matching using stride-k DFA leads to improvement in terms of speedup and latency over traditional DFA matching.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
