GPU-accelerated computation of electron transfer

Siegfried Höfinger, Angela Acocella, Sergiu C. Pop, Tetsu Narumi, Kenji Yasuoka, Titus Beu, Francesco Zerbetto

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Electron transfer is a fundamental process that can be studied with the help of computer simulation. The underlying quantum mechanical description renders the problem a computationally intensive application. In this study, we probe the graphics processing unit (GPU) for suitability to this type of problem. Time-critical components are identified via profiling of an existing implementation and several different variants are tested involving the GPU at increasing levels of abstraction. A publicly available library supporting basic linear algebra operations on the GPU turns out to accelerate the computation approximately 50-fold with minor dependence on actual problem size. The performance gain does not compromise numerical accuracy and is of significant value for practical purposes.

Original languageEnglish
Pages (from-to)2351-2356
Number of pages6
JournalJournal of Computational Chemistry
Volume33
Issue number29
DOIs
Publication statusPublished - 2012 Nov 5

Fingerprint

Electron Transfer
Graphics Processing Unit
Electrons
Numerical Accuracy
Linear algebra
Profiling
Accelerate
Minor
Fold
Probe
Computer Simulation
Computer simulation
Graphics processing unit

Keywords

  • CUBLAS
  • CUDA
  • electron transfer
  • GPU
  • time-dependent Schrödinger equation

ASJC Scopus subject areas

  • Chemistry(all)
  • Computational Mathematics

Cite this

Höfinger, S., Acocella, A., Pop, S. C., Narumi, T., Yasuoka, K., Beu, T., & Zerbetto, F. (2012). GPU-accelerated computation of electron transfer. Journal of Computational Chemistry, 33(29), 2351-2356. https://doi.org/10.1002/jcc.23082

GPU-accelerated computation of electron transfer. / Höfinger, Siegfried; Acocella, Angela; Pop, Sergiu C.; Narumi, Tetsu; Yasuoka, Kenji; Beu, Titus; Zerbetto, Francesco.

In: Journal of Computational Chemistry, Vol. 33, No. 29, 05.11.2012, p. 2351-2356.

Research output: Contribution to journalArticle

Höfinger, S, Acocella, A, Pop, SC, Narumi, T, Yasuoka, K, Beu, T & Zerbetto, F 2012, 'GPU-accelerated computation of electron transfer', Journal of Computational Chemistry, vol. 33, no. 29, pp. 2351-2356. https://doi.org/10.1002/jcc.23082
Höfinger S, Acocella A, Pop SC, Narumi T, Yasuoka K, Beu T et al. GPU-accelerated computation of electron transfer. Journal of Computational Chemistry. 2012 Nov 5;33(29):2351-2356. https://doi.org/10.1002/jcc.23082
Höfinger, Siegfried ; Acocella, Angela ; Pop, Sergiu C. ; Narumi, Tetsu ; Yasuoka, Kenji ; Beu, Titus ; Zerbetto, Francesco. / GPU-accelerated computation of electron transfer. In: Journal of Computational Chemistry. 2012 ; Vol. 33, No. 29. pp. 2351-2356.
@article{93e367488b7941f2b5d7ec7b2906bb3f,
title = "GPU-accelerated computation of electron transfer",
abstract = "Electron transfer is a fundamental process that can be studied with the help of computer simulation. The underlying quantum mechanical description renders the problem a computationally intensive application. In this study, we probe the graphics processing unit (GPU) for suitability to this type of problem. Time-critical components are identified via profiling of an existing implementation and several different variants are tested involving the GPU at increasing levels of abstraction. A publicly available library supporting basic linear algebra operations on the GPU turns out to accelerate the computation approximately 50-fold with minor dependence on actual problem size. The performance gain does not compromise numerical accuracy and is of significant value for practical purposes.",
keywords = "CUBLAS, CUDA, electron transfer, GPU, time-dependent Schr{\"o}dinger equation",
author = "Siegfried H{\"o}finger and Angela Acocella and Pop, {Sergiu C.} and Tetsu Narumi and Kenji Yasuoka and Titus Beu and Francesco Zerbetto",
year = "2012",
month = "11",
day = "5",
doi = "10.1002/jcc.23082",
language = "English",
volume = "33",
pages = "2351--2356",
journal = "Journal of Computational Chemistry",
issn = "0192-8651",
publisher = "John Wiley and Sons Inc.",
number = "29",

}

TY - JOUR

T1 - GPU-accelerated computation of electron transfer

AU - Höfinger, Siegfried

AU - Acocella, Angela

AU - Pop, Sergiu C.

AU - Narumi, Tetsu

AU - Yasuoka, Kenji

AU - Beu, Titus

AU - Zerbetto, Francesco

PY - 2012/11/5

Y1 - 2012/11/5

N2 - Electron transfer is a fundamental process that can be studied with the help of computer simulation. The underlying quantum mechanical description renders the problem a computationally intensive application. In this study, we probe the graphics processing unit (GPU) for suitability to this type of problem. Time-critical components are identified via profiling of an existing implementation and several different variants are tested involving the GPU at increasing levels of abstraction. A publicly available library supporting basic linear algebra operations on the GPU turns out to accelerate the computation approximately 50-fold with minor dependence on actual problem size. The performance gain does not compromise numerical accuracy and is of significant value for practical purposes.

AB - Electron transfer is a fundamental process that can be studied with the help of computer simulation. The underlying quantum mechanical description renders the problem a computationally intensive application. In this study, we probe the graphics processing unit (GPU) for suitability to this type of problem. Time-critical components are identified via profiling of an existing implementation and several different variants are tested involving the GPU at increasing levels of abstraction. A publicly available library supporting basic linear algebra operations on the GPU turns out to accelerate the computation approximately 50-fold with minor dependence on actual problem size. The performance gain does not compromise numerical accuracy and is of significant value for practical purposes.

KW - CUBLAS

KW - CUDA

KW - electron transfer

KW - GPU

KW - time-dependent Schrödinger equation

UR - http://www.scopus.com/inward/record.url?scp=84867083594&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867083594&partnerID=8YFLogxK

U2 - 10.1002/jcc.23082

DO - 10.1002/jcc.23082

M3 - Article

C2 - 22847673

AN - SCOPUS:84867083594

VL - 33

SP - 2351

EP - 2356

JO - Journal of Computational Chemistry

JF - Journal of Computational Chemistry

SN - 0192-8651

IS - 29

ER -