Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes

Dinesh Kaushik, William Gropp, Michael Minkoff, Barry Smith

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

3 Citations (Scopus)

Abstract

Cumulative reaction probability (CRP) calculations provide a viable computational approach to estimate reaction rate coefficients. However, in order to give meaningful results these calculations should be done in many dimensions (ten to fifteen). This makes CRP codes memory intensive. For this reason, these codes use iterative methods to solve the linear systems, where a good fraction of the execution time is spent on matrix-vector multiplication. In this paper, we discuss the tensor product form of applying the system operator on a vector. This approach shows much better performance and provides huge savings in memory as compared to the explicit sparse representation of the system matrix.

Original languageEnglish
Title of host publicationHigh Performance Computing - HiPC 2008 - 15th International Conference, Proceedings
PublisherSpringer Verlag
Pages120-130
Number of pages11
Volume5374 LNCS
ISBN (Print)354089893X, 9783540898931
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event15th International Conference on High Performance Computing, HiPC 2008 - Bangalore, India
Duration: 17 Dec 200820 Dec 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5374 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other15th International Conference on High Performance Computing, HiPC 2008
CountryIndia
CityBangalore
Period17/12/0820/12/08

Fingerprint

Quantum Chemistry
Quantum chemistry
Matrix-vector multiplication
Tensors
Tensor
Data storage equipment
Product Form
Sparse Representation
Reaction Rate
Iterative methods
Tensor Product
Execution Time
Reaction rates
Linear systems
Linear Systems
Iteration
Coefficient
Operator
Estimate

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Kaushik, D., Gropp, W., Minkoff, M., & Smith, B. (2008). Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes. In High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings (Vol. 5374 LNCS, pp. 120-130). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5374 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-540-89894-8_14

Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes. / Kaushik, Dinesh; Gropp, William; Minkoff, Michael; Smith, Barry.

High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings. Vol. 5374 LNCS Springer Verlag, 2008. p. 120-130 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5374 LNCS).

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

Kaushik, D, Gropp, W, Minkoff, M & Smith, B 2008, Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes. in High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings. vol. 5374 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5374 LNCS, Springer Verlag, pp. 120-130, 15th International Conference on High Performance Computing, HiPC 2008, Bangalore, India, 17/12/08. https://doi.org/10.1007/978-3-540-89894-8_14
Kaushik D, Gropp W, Minkoff M, Smith B. Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes. In High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings. Vol. 5374 LNCS. Springer Verlag. 2008. p. 120-130. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-89894-8_14
Kaushik, Dinesh ; Gropp, William ; Minkoff, Michael ; Smith, Barry. / Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes. High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings. Vol. 5374 LNCS Springer Verlag, 2008. pp. 120-130 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c7a6503907524eaf9e4579a806abd85b,
title = "Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes",
abstract = "Cumulative reaction probability (CRP) calculations provide a viable computational approach to estimate reaction rate coefficients. However, in order to give meaningful results these calculations should be done in many dimensions (ten to fifteen). This makes CRP codes memory intensive. For this reason, these codes use iterative methods to solve the linear systems, where a good fraction of the execution time is spent on matrix-vector multiplication. In this paper, we discuss the tensor product form of applying the system operator on a vector. This approach shows much better performance and provides huge savings in memory as compared to the explicit sparse representation of the system matrix.",
author = "Dinesh Kaushik and William Gropp and Michael Minkoff and Barry Smith",
year = "2008",
doi = "10.1007/978-3-540-89894-8_14",
language = "English",
isbn = "354089893X",
volume = "5374 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "120--130",
booktitle = "High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings",

}

TY - GEN

T1 - Improving the performance of tensor matrix vector multiplication in cumulative reaction probability based quantum chemistry codes

AU - Kaushik, Dinesh

AU - Gropp, William

AU - Minkoff, Michael

AU - Smith, Barry

PY - 2008

Y1 - 2008

N2 - Cumulative reaction probability (CRP) calculations provide a viable computational approach to estimate reaction rate coefficients. However, in order to give meaningful results these calculations should be done in many dimensions (ten to fifteen). This makes CRP codes memory intensive. For this reason, these codes use iterative methods to solve the linear systems, where a good fraction of the execution time is spent on matrix-vector multiplication. In this paper, we discuss the tensor product form of applying the system operator on a vector. This approach shows much better performance and provides huge savings in memory as compared to the explicit sparse representation of the system matrix.

AB - Cumulative reaction probability (CRP) calculations provide a viable computational approach to estimate reaction rate coefficients. However, in order to give meaningful results these calculations should be done in many dimensions (ten to fifteen). This makes CRP codes memory intensive. For this reason, these codes use iterative methods to solve the linear systems, where a good fraction of the execution time is spent on matrix-vector multiplication. In this paper, we discuss the tensor product form of applying the system operator on a vector. This approach shows much better performance and provides huge savings in memory as compared to the explicit sparse representation of the system matrix.

UR - http://www.scopus.com/inward/record.url?scp=58449097645&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58449097645&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-89894-8_14

DO - 10.1007/978-3-540-89894-8_14

M3 - Conference contribution

AN - SCOPUS:58449097645

SN - 354089893X

SN - 9783540898931

VL - 5374 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 120

EP - 130

BT - High Performance Computing - HiPC 2008 - 15th International Conference, Proceedings

PB - Springer Verlag

ER -