Deployment on GPUs of an application in computational atomic physics

P. Fortin; R. Habel; F. Jézéquel; J.-L. Lamotte; N.S. Scott

doi:10.1109/IPDPS.2011.285

Deployment on GPUs of an application in computational atomic physics

P. Fortin, R. Habel, F. Jézéquel, J.-L. Lamotte, N.S. Scott

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

2 Citations (Scopus)

Abstract

This paper describes the deployment on GPUs of PROP, a program of the 2DRMP suite which models electron collisions with H-like atoms and ions. Because performance on GPUs is better in single precision than in double precision, the numerical stability of the PROP program in single precision has been studied. The numerical quality of PROP results computed in single precision and their impact on the next program of the 2DRMP suite has been analyzed. Successive versions of the PROP program on GPUs have been developed in order to improve its performance. Particular attention has been paid to the optimization of data transfers and of linear algebra operations. Performance obtained on several architectures (including NVIDIA Fermi) are presented.

Original language	English
Title of host publication	25th IEEE International Symposium on Parallel and Distributed Processing
Subtitle of host publication	IPDPS 2011
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1359-1366
Number of pages	8
ISBN (Electronic)	978-0-7695-4577-6
ISBN (Print)	978-1-61284-425-1
DOIs	https://doi.org/10.1109/IPDPS.2011.285
Publication status	Published - May 2011
Event	12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11) - Anchorage, Alaska, United States Duration: 16 May 2011 → 20 May 2011

Conference

Conference	12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11)
Country	United States
City	Anchorage, Alaska
Period	16/05/2011 → 20/05/2011

Bibliographical note

ISSN: 1530-2075

ASJC Scopus subject areas

Computational Theory and Mathematics
Software
Theoretical Computer Science

Access to Document

10.1109/IPDPS.2011.285

Fingerprint Dive into the research topics of 'Deployment on GPUs of an application in computational atomic physics'. Together they form a unique fingerprint.

Cite this

@inproceedings{cc73aba0e0a14617ad8400e5931f0ab2,

title = "Deployment on GPUs of an application in computational atomic physics",

abstract = "This paper describes the deployment on GPUs of PROP, a program of the 2DRMP suite which models electron collisions with H-like atoms and ions. Because performance on GPUs is better in single precision than in double precision, the numerical stability of the PROP program in single precision has been studied. The numerical quality of PROP results computed in single precision and their impact on the next program of the 2DRMP suite has been analyzed. Successive versions of the PROP program on GPUs have been developed in order to improve its performance. Particular attention has been paid to the optimization of data transfers and of linear algebra operations. Performance obtained on several architectures (including NVIDIA Fermi) are presented.",

author = "P. Fortin and R. Habel and F. J{\'e}z{\'e}quel and J.-L. Lamotte and N.S. Scott",

note = "ISSN: 1530-2075; 12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11) ; Conference date: 16-05-2011 Through 20-05-2011",

year = "2011",

month = may,

doi = "10.1109/IPDPS.2011.285",

language = "English",

isbn = "978-1-61284-425-1",

pages = "1359--1366",

booktitle = "25th IEEE International Symposium on Parallel and Distributed Processing",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

}

Fortin, P, Habel, R, Jézéquel, F, Lamotte, J-L & Scott, NS 2011, Deployment on GPUs of an application in computational atomic physics. in 25th IEEE International Symposium on Parallel and Distributed Processing: IPDPS 2011. Institute of Electrical and Electronics Engineers (IEEE), pp. 1359-1366, 12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11), Anchorage, Alaska, United States, 16/05/2011. https://doi.org/10.1109/IPDPS.2011.285

Deployment on GPUs of an application in computational atomic physics. / Fortin, P.; Habel, R.; Jézéquel, F.; Lamotte, J.-L.; Scott, N.S.

25th IEEE International Symposium on Parallel and Distributed Processing: IPDPS 2011. Institute of Electrical and Electronics Engineers (IEEE), 2011. p. 1359-1366.

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

TY - GEN

T1 - Deployment on GPUs of an application in computational atomic physics

AU - Fortin, P.

AU - Habel, R.

AU - Jézéquel, F.

AU - Lamotte, J.-L.

AU - Scott, N.S.

N1 - ISSN: 1530-2075

PY - 2011/5

Y1 - 2011/5

N2 - This paper describes the deployment on GPUs of PROP, a program of the 2DRMP suite which models electron collisions with H-like atoms and ions. Because performance on GPUs is better in single precision than in double precision, the numerical stability of the PROP program in single precision has been studied. The numerical quality of PROP results computed in single precision and their impact on the next program of the 2DRMP suite has been analyzed. Successive versions of the PROP program on GPUs have been developed in order to improve its performance. Particular attention has been paid to the optimization of data transfers and of linear algebra operations. Performance obtained on several architectures (including NVIDIA Fermi) are presented.

AB - This paper describes the deployment on GPUs of PROP, a program of the 2DRMP suite which models electron collisions with H-like atoms and ions. Because performance on GPUs is better in single precision than in double precision, the numerical stability of the PROP program in single precision has been studied. The numerical quality of PROP results computed in single precision and their impact on the next program of the 2DRMP suite has been analyzed. Successive versions of the PROP program on GPUs have been developed in order to improve its performance. Particular attention has been paid to the optimization of data transfers and of linear algebra operations. Performance obtained on several architectures (including NVIDIA Fermi) are presented.

U2 - 10.1109/IPDPS.2011.285

DO - 10.1109/IPDPS.2011.285

M3 - Conference contribution

SN - 978-1-61284-425-1

SP - 1359

EP - 1366

BT - 25th IEEE International Symposium on Parallel and Distributed Processing

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11)

Y2 - 16 May 2011 through 20 May 2011

ER -