Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster

Thomas Hauser; Mark A. Perl; Barton L. Smith

doi:10.1115/FEDSM2007-37649

Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster

Thomas Hauser, Mark A. Perl, Barton L. Smith

Utah State University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we present our design of a portable cluster supercomputer created specifically for processing PIV image data. To make the system portable for a laboratory environment, the cluster computer is required to run on minimal power and be light weight. The final cluster design was based on the Intel®Quad-Core Xeon®X3210 low-power processors. This hardware setup allows for a total system consisting of twelve processing nodes with a total of 48 CPU cores, one master node with 5 TB of storage, and a Gigabit Ethernet connection. The total configuration along with the case has an approximate gross weight of 390 lbs and works well on one standard 120-V, 20-Amp electric circuit. With this cluster computer, a speedup of about 28 relative to standard serial processing of a PIV data set could be achieved.

Original language	English
Title of host publication	2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007
Pages	187-194
Number of pages	8
Edition	PART A
DOIs	https://doi.org/10.1115/FEDSM2007-37649
State	Published - 2007
Event	2007 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007 - San Diego, CA, United States Duration: Jul 30 2007 → Aug 2 2007

Publication series

Name	2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007
Number	PART A
Volume	2 FORA

Conference

Conference	2007 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007
Country/Territory	United States
City	San Diego, CA
Period	07/30/07 → 08/2/07

Access to Document

10.1115/FEDSM2007-37649

Cite this

Hauser, T., Perl, M. A., & Smith, B. L. (2007). Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster. In 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007 (PART A ed., pp. 187-194). (2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007; Vol. 2 FORA, No. PART A). https://doi.org/10.1115/FEDSM2007-37649

Hauser, Thomas ; Perl, Mark A. ; Smith, Barton L. / Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster. 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007. PART A. ed. 2007. pp. 187-194 (2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007; PART A).

@inproceedings{418c7722f5ee43a382a14b9aca2228cb,

title = "Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster",

abstract = "In this paper, we present our design of a portable cluster supercomputer created specifically for processing PIV image data. To make the system portable for a laboratory environment, the cluster computer is required to run on minimal power and be light weight. The final cluster design was based on the Intel{\textregistered}Quad-Core Xeon{\textregistered}X3210 low-power processors. This hardware setup allows for a total system consisting of twelve processing nodes with a total of 48 CPU cores, one master node with 5 TB of storage, and a Gigabit Ethernet connection. The total configuration along with the case has an approximate gross weight of 390 lbs and works well on one standard 120-V, 20-Amp electric circuit. With this cluster computer, a speedup of about 28 relative to standard serial processing of a PIV data set could be achieved.",

author = "Thomas Hauser and Perl, \{Mark A.\} and Smith, \{Barton L.\}",

year = "2007",

doi = "10.1115/FEDSM2007-37649",

language = "English",

isbn = "0791842886",

series = "2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007",

number = "PART A",

pages = "187--194",

booktitle = "2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007",

edition = "PART A",

note = "2007 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007 ; Conference date: 30-07-2007 Through 02-08-2007",

}

Hauser, T, Perl, MA & Smith, BL 2007, Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster. in 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007. PART A edn, 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007, no. PART A, vol. 2 FORA, pp. 187-194, 2007 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007, San Diego, CA, United States, 07/30/07. https://doi.org/10.1115/FEDSM2007-37649

Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster. / Hauser, Thomas; Perl, Mark A.; Smith, Barton L.
2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007. PART A. ed. 2007. p. 187-194 (2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007; Vol. 2 FORA, No. PART A).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster

AU - Hauser, Thomas

AU - Perl, Mark A.

AU - Smith, Barton L.

PY - 2007

Y1 - 2007

N2 - In this paper, we present our design of a portable cluster supercomputer created specifically for processing PIV image data. To make the system portable for a laboratory environment, the cluster computer is required to run on minimal power and be light weight. The final cluster design was based on the Intel®Quad-Core Xeon®X3210 low-power processors. This hardware setup allows for a total system consisting of twelve processing nodes with a total of 48 CPU cores, one master node with 5 TB of storage, and a Gigabit Ethernet connection. The total configuration along with the case has an approximate gross weight of 390 lbs and works well on one standard 120-V, 20-Amp electric circuit. With this cluster computer, a speedup of about 28 relative to standard serial processing of a PIV data set could be achieved.

AB - In this paper, we present our design of a portable cluster supercomputer created specifically for processing PIV image data. To make the system portable for a laboratory environment, the cluster computer is required to run on minimal power and be light weight. The final cluster design was based on the Intel®Quad-Core Xeon®X3210 low-power processors. This hardware setup allows for a total system consisting of twelve processing nodes with a total of 48 CPU cores, one master node with 5 TB of storage, and a Gigabit Ethernet connection. The total configuration along with the case has an approximate gross weight of 390 lbs and works well on one standard 120-V, 20-Amp electric circuit. With this cluster computer, a speedup of about 28 relative to standard serial processing of a PIV data set could be achieved.

UR - https://www.scopus.com/pages/publications/40449124795

U2 - 10.1115/FEDSM2007-37649

DO - 10.1115/FEDSM2007-37649

M3 - Conference contribution

AN - SCOPUS:40449124795

SN - 0791842886

SN - 9780791842881

T3 - 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007

SP - 187

EP - 194

BT - 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007

T2 - 2007 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007

Y2 - 30 July 2007 through 2 August 2007

ER -

Hauser T, Perl MA, Smith BL. Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster. In 2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007. PART A ed. 2007. p. 187-194. (2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007; PART A). doi: 10.1115/FEDSM2007-37649

Distributed processing of high-speed stereo particle image velocimetry data with a low power beowulf cluster

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this