A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

5 Citations (Scopus)

Abstract

Pattern Matching is a computationally intensive task used in many research fields and real world applications. Due to the ever-growing volume of data to be processed, and increasing link speeds, the number of patterns to be matched has risen significantly. In this paper we explore the parallel capabilities of modern General Purpose Graphics Processing Units (GPGPU) applications for high speed pattern matching. A highly compressed failure-less Aho-Corasick algorithm is presented for Intrusion Detection Systems on off-the-shelf hardware. This approach maximises the bandwidth for data transfers between the host and the Graphics Processing Unit (GPU). Experiments are performed on multiple alphabet sizes, demonstrating the capabilities of the library to be used in different research fields, while sustaining an adequate throughput for intrusion detection systems or DNA sequencing. The work also explores the performance impact of adequate prefix matching for alphabet sizes and varying pattern numbers achieving speeds up to 8Gbps and low memory consumption for intrusion detection systems.
LanguageEnglish
Title of host publicationSIN '14 Proceedings of the 7th International Conference on Security of Information and Networks
Place of PublicationNew York
DOIs
Publication statusPublished - Sep 2014
Event7th International Conference on the Security of Information and Networks (SIN14) - University of Glasgow, Glasgow, United Kingdom
Duration: 9 Sep 201411 Sep 2014

Conference

Conference7th International Conference on the Security of Information and Networks (SIN14)
CountryUnited Kingdom
CityGlasgow
Period9/09/1411/09/14

Fingerprint

Intrusion detection
Pattern matching
Data storage equipment
On-off control systems
Data transfer
Computer hardware
DNA
Throughput
Bandwidth
Graphics processing unit
Experiments

Keywords

  • pattern matching
  • high speed pattern matching
  • Aho-Corasick algorithm
  • information networks

Cite this

Bellekens, X. J. A., Tachtatzis, C., Atkinson, R. C., Renfrew, C., & Kirkham, T. (2014). A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems. In SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks New York. https://doi.org/10.1145/2659651.2659723
Bellekens, Xavier J.A. ; Tachtatzis, Christos ; Atkinson, Robert C. ; Renfrew, Craig ; Kirkham, Tony. / A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems. SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks . New York, 2014.
@inproceedings{3f1fea6504644ac79f8ab83808e120ca,
title = "A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems",
abstract = "Pattern Matching is a computationally intensive task used in many research fields and real world applications. Due to the ever-growing volume of data to be processed, and increasing link speeds, the number of patterns to be matched has risen significantly. In this paper we explore the parallel capabilities of modern General Purpose Graphics Processing Units (GPGPU) applications for high speed pattern matching. A highly compressed failure-less Aho-Corasick algorithm is presented for Intrusion Detection Systems on off-the-shelf hardware. This approach maximises the bandwidth for data transfers between the host and the Graphics Processing Unit (GPU). Experiments are performed on multiple alphabet sizes, demonstrating the capabilities of the library to be used in different research fields, while sustaining an adequate throughput for intrusion detection systems or DNA sequencing. The work also explores the performance impact of adequate prefix matching for alphabet sizes and varying pattern numbers achieving speeds up to 8Gbps and low memory consumption for intrusion detection systems.",
keywords = "pattern matching, high speed pattern matching, Aho-Corasick algorithm, information networks",
author = "Bellekens, {Xavier J.A.} and Christos Tachtatzis and Atkinson, {Robert C.} and Craig Renfrew and Tony Kirkham",
year = "2014",
month = "9",
doi = "10.1145/2659651.2659723",
language = "English",
isbn = "978-1-4503-3033-6",
booktitle = "SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks",

}

Bellekens, XJA, Tachtatzis, C, Atkinson, RC, Renfrew, C & Kirkham, T 2014, A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems. in SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks . New York, 7th International Conference on the Security of Information and Networks (SIN14), Glasgow, United Kingdom, 9/09/14. https://doi.org/10.1145/2659651.2659723

A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems. / Bellekens, Xavier J.A.; Tachtatzis, Christos; Atkinson, Robert C.; Renfrew, Craig; Kirkham, Tony.

SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks . New York, 2014.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems

AU - Bellekens, Xavier J.A.

AU - Tachtatzis, Christos

AU - Atkinson, Robert C.

AU - Renfrew, Craig

AU - Kirkham, Tony

PY - 2014/9

Y1 - 2014/9

N2 - Pattern Matching is a computationally intensive task used in many research fields and real world applications. Due to the ever-growing volume of data to be processed, and increasing link speeds, the number of patterns to be matched has risen significantly. In this paper we explore the parallel capabilities of modern General Purpose Graphics Processing Units (GPGPU) applications for high speed pattern matching. A highly compressed failure-less Aho-Corasick algorithm is presented for Intrusion Detection Systems on off-the-shelf hardware. This approach maximises the bandwidth for data transfers between the host and the Graphics Processing Unit (GPU). Experiments are performed on multiple alphabet sizes, demonstrating the capabilities of the library to be used in different research fields, while sustaining an adequate throughput for intrusion detection systems or DNA sequencing. The work also explores the performance impact of adequate prefix matching for alphabet sizes and varying pattern numbers achieving speeds up to 8Gbps and low memory consumption for intrusion detection systems.

AB - Pattern Matching is a computationally intensive task used in many research fields and real world applications. Due to the ever-growing volume of data to be processed, and increasing link speeds, the number of patterns to be matched has risen significantly. In this paper we explore the parallel capabilities of modern General Purpose Graphics Processing Units (GPGPU) applications for high speed pattern matching. A highly compressed failure-less Aho-Corasick algorithm is presented for Intrusion Detection Systems on off-the-shelf hardware. This approach maximises the bandwidth for data transfers between the host and the Graphics Processing Unit (GPU). Experiments are performed on multiple alphabet sizes, demonstrating the capabilities of the library to be used in different research fields, while sustaining an adequate throughput for intrusion detection systems or DNA sequencing. The work also explores the performance impact of adequate prefix matching for alphabet sizes and varying pattern numbers achieving speeds up to 8Gbps and low memory consumption for intrusion detection systems.

KW - pattern matching

KW - high speed pattern matching

KW - Aho-Corasick algorithm

KW - information networks

U2 - 10.1145/2659651.2659723

DO - 10.1145/2659651.2659723

M3 - Conference contribution book

SN - 978-1-4503-3033-6

BT - SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks

CY - New York

ER -

Bellekens XJA, Tachtatzis C, Atkinson RC, Renfrew C, Kirkham T. A highly-efficient memory-compression scheme for GPU-accelerated intrusion detection systems. In SIN '14 Proceedings of the 7th International Conference on Security of Information and Networks . New York. 2014 https://doi.org/10.1145/2659651.2659723