Using multiple GPUs to accelerate string searching for digital forensic analysis

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

68 Downloads (Pure)

Abstract

String searching within a large corpus of data is an important component of digital forensic (DF) analysis techniques such as file carving. The continuing increase in capacity of consumer storage devices requires corresponding im-provements to the performance of string searching techniques. As string search-ing is a trivially-parallelisable problem, GPGPU approaches are a natural fit – but previous studies have found that local storage presents an insurmountable performance bottleneck. We show that this need not be the case with modern hardware, and demonstrate substantial performance improvements from the use of single and multiple GPUs when searching for strings within a typical forensic disk image.
Original languageEnglish
Title of host publicationSADFE Proceedings 2016
PublisherAssociation of Digital Forensics, Security and Law (ADFSL)
Pages67-78
Number of pages12
Publication statusPublished - Sep 2016

Fingerprint

Hardware
Graphics processing unit
Digital forensics

Cite this

Bayne, E., Ferguson, R. I., Sampson, A. T., & Isaacs, J. P. (2016). Using multiple GPUs to accelerate string searching for digital forensic analysis. In SADFE Proceedings 2016 (pp. 67-78). Association of Digital Forensics, Security and Law (ADFSL).
Bayne, Ethan ; Ferguson, R. Ian ; Sampson, Adam T. ; Isaacs, John P. / Using multiple GPUs to accelerate string searching for digital forensic analysis. SADFE Proceedings 2016. Association of Digital Forensics, Security and Law (ADFSL), 2016. pp. 67-78
@inproceedings{40c9db8df41d48e6bd16f803ec1c3a5f,
title = "Using multiple GPUs to accelerate string searching for digital forensic analysis",
abstract = "String searching within a large corpus of data is an important component of digital forensic (DF) analysis techniques such as file carving. The continuing increase in capacity of consumer storage devices requires corresponding im-provements to the performance of string searching techniques. As string search-ing is a trivially-parallelisable problem, GPGPU approaches are a natural fit – but previous studies have found that local storage presents an insurmountable performance bottleneck. We show that this need not be the case with modern hardware, and demonstrate substantial performance improvements from the use of single and multiple GPUs when searching for strings within a typical forensic disk image.",
author = "Ethan Bayne and Ferguson, {R. Ian} and Sampson, {Adam T.} and Isaacs, {John P.}",
year = "2016",
month = "9",
language = "English",
pages = "67--78",
booktitle = "SADFE Proceedings 2016",
publisher = "Association of Digital Forensics, Security and Law (ADFSL)",

}

Bayne, E, Ferguson, RI, Sampson, AT & Isaacs, JP 2016, Using multiple GPUs to accelerate string searching for digital forensic analysis. in SADFE Proceedings 2016. Association of Digital Forensics, Security and Law (ADFSL), pp. 67-78.

Using multiple GPUs to accelerate string searching for digital forensic analysis. / Bayne, Ethan; Ferguson, R. Ian; Sampson, Adam T.; Isaacs, John P.

SADFE Proceedings 2016. Association of Digital Forensics, Security and Law (ADFSL), 2016. p. 67-78.

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

TY - GEN

T1 - Using multiple GPUs to accelerate string searching for digital forensic analysis

AU - Bayne, Ethan

AU - Ferguson, R. Ian

AU - Sampson, Adam T.

AU - Isaacs, John P.

PY - 2016/9

Y1 - 2016/9

N2 - String searching within a large corpus of data is an important component of digital forensic (DF) analysis techniques such as file carving. The continuing increase in capacity of consumer storage devices requires corresponding im-provements to the performance of string searching techniques. As string search-ing is a trivially-parallelisable problem, GPGPU approaches are a natural fit – but previous studies have found that local storage presents an insurmountable performance bottleneck. We show that this need not be the case with modern hardware, and demonstrate substantial performance improvements from the use of single and multiple GPUs when searching for strings within a typical forensic disk image.

AB - String searching within a large corpus of data is an important component of digital forensic (DF) analysis techniques such as file carving. The continuing increase in capacity of consumer storage devices requires corresponding im-provements to the performance of string searching techniques. As string search-ing is a trivially-parallelisable problem, GPGPU approaches are a natural fit – but previous studies have found that local storage presents an insurmountable performance bottleneck. We show that this need not be the case with modern hardware, and demonstrate substantial performance improvements from the use of single and multiple GPUs when searching for strings within a typical forensic disk image.

M3 - Conference contribution

SP - 67

EP - 78

BT - SADFE Proceedings 2016

PB - Association of Digital Forensics, Security and Law (ADFSL)

ER -

Bayne E, Ferguson RI, Sampson AT, Isaacs JP. Using multiple GPUs to accelerate string searching for digital forensic analysis. In SADFE Proceedings 2016. Association of Digital Forensics, Security and Law (ADFSL). 2016. p. 67-78