Skip to main navigation Skip to search Skip to main content

Towards a Fast Regular Expression Matching Method over Compressed Traffic

  • Xiuwen Sun
  • , Hao Li
  • , Xingxing Lu
  • , Dan Zhao
  • , Zheng Peng
  • , Chengchen Hu
  • Xi'an Jiaotong University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Nowadays, Deep Packet Inspection (DPI) becomes a critical component of the network traffic detection applications. For comprehensive analysis of traffic, regular expression matching as the core technique of DPI is widely used. However, web services tend to compress their traffic for less data transmission, which challenges the regular expression matching to achieve wire-speed processing. In this paper, we propose Twins, a fast regular expression matching method over compressed traffic that leverages the returned states encoding in the compression to skip the bytes to be scanned. In our evaluation results, Twins can skip about 90% compression data and can achieve 1.5Gbps throughput, which gains 2.7∼3.4 performance boost to the state-of-the-art work.

Original languageEnglish
Title of host publication2018 IEEE/ACM 26th International Symposium on Quality of Service, IWQoS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538625422
DOIs
StatePublished - 22 Jan 2019
Event26th IEEE/ACM International Symposium on Quality of Service, IWQoS 2018 - Banff, Canada
Duration: 4 Jun 20186 Jun 2018

Publication series

Name2018 IEEE/ACM 26th International Symposium on Quality of Service, IWQoS 2018

Conference

Conference26th IEEE/ACM International Symposium on Quality of Service, IWQoS 2018
Country/TerritoryCanada
CityBanff
Period4/06/186/06/18

Keywords

  • Compressed Traffic
  • Deep Packet Inspection
  • Multi-Pattern Matching
  • Regular Expression Matching

Fingerprint

Dive into the research topics of 'Towards a Fast Regular Expression Matching Method over Compressed Traffic'. Together they form a unique fingerprint.

Cite this