High-cost CFD on a low-cost cluster

Thomas Hauser; Timothy I. Mattox; Raymond P. LeBeau; Henry G. Dietz; P. George Huang

doi:10.1109/SC.2000.10020

High-cost CFD on a low-cost cluster

Thomas Hauser, Timothy I. Mattox, Raymond P. LeBeau, Henry G. Dietz, P. George Huang

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

11 Scopus citations

Abstract

Direct numerical simulation of the Navier-Stokes equations (DNS) is an important technique for the future of computational fluid dynamics (CFD) in engineering applications. However, DNS requires massive computing resources. This paper presents a new approach for implementing high-cost DNS CFD using low-cost cluster hardware. After describing the DNS CFD code DNSTool, the paper focuses on the techniques and tools that we have developed to customize the performance of a cluster implementation of this application. This tuning of system performance involves both recoding of the application and careful engineering of the cluster design. Using the cluster KLAT2 (Kentucky Linux Athlon Testbed 2), while DNSTool cannot match the $0.64 per MFLOPS that KLAT2 achieves on single precision ScaLAPACK, it is very efficient; DNSTool on KLAT2 achieves price/performance of $2.75 per MFLOPS double precision and $1.86 single precision. Further, the code and tools are all, or will soon be, made freely available as full source code.

Original language	English
Title of host publication	SC 2000 - Proceedings of the 2000 ACM/IEEE Conference on Supercomputing
ISBN (Electronic)	0780398025
DOIs	https://doi.org/10.1109/SC.2000.10020
State	Published - 2000
Event	2000 ACM/IEEE Conference on Supercomputing, SC 2000 - Dallas, United States Duration: Nov 4 2000 → Nov 10 2000

Publication series

Name	Proceedings of the International Conference on Supercomputing
Volume	2000-November

Conference

Conference	2000 ACM/IEEE Conference on Supercomputing, SC 2000
Country/Territory	United States
City	Dallas
Period	11/4/00 → 11/10/00

Bibliographical note

Funding Information:
Thomas Hauser is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center and the Center for Computational Science at the University of Kentucky. Ray LeBeau is supported by Kentucky NASA-EPSCoR. George Huang is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center. The initial work on DNSTool was supported by SFB 255 at the University of Technology Munich under Rainer Friedrich. Special thanks to AMD for their generous contributions of the Athlon processors for KLAT2 and to the many students who got their first taste of cluster supercomputing amongst the rubble of discarded packing material from which KLAT2 rose on April 11, 2000.

Publisher Copyright:
© 2000 IEEE.

ASJC Scopus subject areas

Computer Science (all)

Access to Document

10.1109/SC.2000.10020

Cite this

@inproceedings{d107c7a93f574642bb7686a4b56be881,

title = "High-cost CFD on a low-cost cluster",

abstract = "Direct numerical simulation of the Navier-Stokes equations (DNS) is an important technique for the future of computational fluid dynamics (CFD) in engineering applications. However, DNS requires massive computing resources. This paper presents a new approach for implementing high-cost DNS CFD using low-cost cluster hardware. After describing the DNS CFD code DNSTool, the paper focuses on the techniques and tools that we have developed to customize the performance of a cluster implementation of this application. This tuning of system performance involves both recoding of the application and careful engineering of the cluster design. Using the cluster KLAT2 (Kentucky Linux Athlon Testbed 2), while DNSTool cannot match the $0.64 per MFLOPS that KLAT2 achieves on single precision ScaLAPACK, it is very efficient; DNSTool on KLAT2 achieves price/performance of $2.75 per MFLOPS double precision and $1.86 single precision. Further, the code and tools are all, or will soon be, made freely available as full source code.",

author = "Thomas Hauser and Mattox, {Timothy I.} and LeBeau, {Raymond P.} and Dietz, {Henry G.} and Huang, {P. George}",

note = "Funding Information: Thomas Hauser is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center and the Center for Computational Science at the University of Kentucky. Ray LeBeau is supported by Kentucky NASA-EPSCoR. George Huang is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center. The initial work on DNSTool was supported by SFB 255 at the University of Technology Munich under Rainer Friedrich. Special thanks to AMD for their generous contributions of the Athlon processors for KLAT2 and to the many students who got their first taste of cluster supercomputing amongst the rubble of discarded packing material from which KLAT2 rose on April 11, 2000. Publisher Copyright: {\textcopyright} 2000 IEEE.; 2000 ACM/IEEE Conference on Supercomputing, SC 2000 ; Conference date: 04-11-2000 Through 10-11-2000",

year = "2000",

doi = "10.1109/SC.2000.10020",

language = "English",

series = "Proceedings of the International Conference on Supercomputing",

booktitle = "SC 2000 - Proceedings of the 2000 ACM/IEEE Conference on Supercomputing",

}

Hauser, T, Mattox, TI, LeBeau, RP, Dietz, HG & Huang, PG 2000, High-cost CFD on a low-cost cluster. in SC 2000 - Proceedings of the 2000 ACM/IEEE Conference on Supercomputing. Proceedings of the International Conference on Supercomputing, vol. 2000-November, 2000 ACM/IEEE Conference on Supercomputing, SC 2000, Dallas, United States, 11/4/00. https://doi.org/10.1109/SC.2000.10020

TY - GEN

T1 - High-cost CFD on a low-cost cluster

AU - Hauser, Thomas

AU - Mattox, Timothy I.

AU - LeBeau, Raymond P.

AU - Dietz, Henry G.

AU - Huang, P. George

N1 - Funding Information: Thomas Hauser is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center and the Center for Computational Science at the University of Kentucky. Ray LeBeau is supported by Kentucky NASA-EPSCoR. George Huang is supported by NASA grant NAG3-2099, sponsored by the NASA-Glenn Research Center. The initial work on DNSTool was supported by SFB 255 at the University of Technology Munich under Rainer Friedrich. Special thanks to AMD for their generous contributions of the Athlon processors for KLAT2 and to the many students who got their first taste of cluster supercomputing amongst the rubble of discarded packing material from which KLAT2 rose on April 11, 2000. Publisher Copyright: © 2000 IEEE.

PY - 2000

Y1 - 2000

N2 - Direct numerical simulation of the Navier-Stokes equations (DNS) is an important technique for the future of computational fluid dynamics (CFD) in engineering applications. However, DNS requires massive computing resources. This paper presents a new approach for implementing high-cost DNS CFD using low-cost cluster hardware. After describing the DNS CFD code DNSTool, the paper focuses on the techniques and tools that we have developed to customize the performance of a cluster implementation of this application. This tuning of system performance involves both recoding of the application and careful engineering of the cluster design. Using the cluster KLAT2 (Kentucky Linux Athlon Testbed 2), while DNSTool cannot match the $0.64 per MFLOPS that KLAT2 achieves on single precision ScaLAPACK, it is very efficient; DNSTool on KLAT2 achieves price/performance of $2.75 per MFLOPS double precision and $1.86 single precision. Further, the code and tools are all, or will soon be, made freely available as full source code.

AB - Direct numerical simulation of the Navier-Stokes equations (DNS) is an important technique for the future of computational fluid dynamics (CFD) in engineering applications. However, DNS requires massive computing resources. This paper presents a new approach for implementing high-cost DNS CFD using low-cost cluster hardware. After describing the DNS CFD code DNSTool, the paper focuses on the techniques and tools that we have developed to customize the performance of a cluster implementation of this application. This tuning of system performance involves both recoding of the application and careful engineering of the cluster design. Using the cluster KLAT2 (Kentucky Linux Athlon Testbed 2), while DNSTool cannot match the $0.64 per MFLOPS that KLAT2 achieves on single precision ScaLAPACK, it is very efficient; DNSTool on KLAT2 achieves price/performance of $2.75 per MFLOPS double precision and $1.86 single precision. Further, the code and tools are all, or will soon be, made freely available as full source code.

UR - http://www.scopus.com/inward/record.url?scp=85117228879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85117228879&partnerID=8YFLogxK

U2 - 10.1109/SC.2000.10020

DO - 10.1109/SC.2000.10020

M3 - Conference contribution

AN - SCOPUS:85117228879

T3 - Proceedings of the International Conference on Supercomputing

BT - SC 2000 - Proceedings of the 2000 ACM/IEEE Conference on Supercomputing

T2 - 2000 ACM/IEEE Conference on Supercomputing, SC 2000

Y2 - 4 November 2000 through 10 November 2000

ER -

High-cost CFD on a low-cost cluster

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this