Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU

Andrés Rodríguez-Hernandez; Armando M. Gómez-Torres; Edmundo del Valle-Gallegos; Javier Jimenez-Escalante; Nico Trost; Victor H. Sanchez-Espinoza

doi:10.1007/978-3-319-32243-8_29

Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU

Andrés Rodríguez-Hernandez, Armando M. Gómez-Torres, Edmundo del Valle-Gallegos, Javier Jimenez-Escalante, Nico Trost, Victor H. Sanchez-Espinoza

Escuela Superior de Física y Matemáticas (ESFM)

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

Abstract

This paper presents the results of the accelerated solution of the linear algebraic system Av = b arising from a nodal finite element method implemented in the neutron diffusion code AZKIND to solve 3D problems. The numerical solution of full nuclear reactor cores with AZKIND implies the generation of large sparse algebraic systems that produce bottle-necks in the iterative solution. Aiming to alleviate the overload of the algorithm, an acceleration technique has to be implemented. Consequently, a Fortran plug-in of the open source linear algebra PARALUTION library (C ++) was integrated into the AZKIND source code (Fortran 95). This implementation allows AZKIND to use GPU_s as well as CPUs, threading into the GPU thousands of arithmetic operations for parallel processing. Selected examples of preliminary investigations performed for a cluster of nuclear fuel assemblies are presented and the obtained results are discussed in this paper.

Original language	English
Title of host publication	High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers
Editors	Isidoro Gitler, Jaime Klapp
Publisher	Springer Verlag
Pages	419-431
Number of pages	13
ISBN (Print)	9783319322421
DOIs	https://doi.org/10.1007/978-3-319-32243-8_29
State	Published - 2016
Event	6th International Conference on High Performance Computer Applications, ISUM 2015 - Mexico City, Mexico Duration: 9 Mar 2015 → 13 Mar 2015

Publication series

Name	Communications in Computer and Information Science
Volume	595
ISSN (Print)	1865-0929

Conference

Conference	6th International Conference on High Performance Computer Applications, ISUM 2015
Country/Territory	Mexico
City	Mexico City
Period	9/03/15 → 13/03/15

Access to Document

10.1007/978-3-319-32243-8_29

Cite this

Rodríguez-Hernandez, A., Gómez-Torres, A. M., del Valle-Gallegos, E., Jimenez-Escalante, J., Trost, N., & Sanchez-Espinoza, V. H. (2016). Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU. In I. Gitler, & J. Klapp (Eds.), High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers (pp. 419-431). (Communications in Computer and Information Science; Vol. 595). Springer Verlag. https://doi.org/10.1007/978-3-319-32243-8_29

Rodríguez-Hernandez, Andrés ; Gómez-Torres, Armando M. ; del Valle-Gallegos, Edmundo et al. / Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU. High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers. editor / Isidoro Gitler ; Jaime Klapp. Springer Verlag, 2016. pp. 419-431 (Communications in Computer and Information Science).

@inproceedings{9d7a5b1b5a4c4ebd9424af9922234347,

title = "Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU",

abstract = "This paper presents the results of the accelerated solution of the linear algebraic system Av = b arising from a nodal finite element method implemented in the neutron diffusion code AZKIND to solve 3D problems. The numerical solution of full nuclear reactor cores with AZKIND implies the generation of large sparse algebraic systems that produce bottle-necks in the iterative solution. Aiming to alleviate the overload of the algorithm, an acceleration technique has to be implemented. Consequently, a Fortran plug-in of the open source linear algebra PARALUTION library (C ++) was integrated into the AZKIND source code (Fortran 95). This implementation allows AZKIND to use GPUs as well as CPUs, threading into the GPU thousands of arithmetic operations for parallel processing. Selected examples of preliminary investigations performed for a cluster of nuclear fuel assemblies are presented and the obtained results are discussed in this paper.",

author = "Andr{\'e}s Rodr{\'i}guez-Hernandez and G{\'o}mez-Torres, {Armando M.} and {del Valle-Gallegos}, Edmundo and Javier Jimenez-Escalante and Nico Trost and Sanchez-Espinoza, {Victor H.}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; 6th International Conference on High Performance Computer Applications, ISUM 2015 ; Conference date: 09-03-2015 Through 13-03-2015",

year = "2016",

doi = "10.1007/978-3-319-32243-8_29",

language = "Ingl{\'e}s",

isbn = "9783319322421",

series = "Communications in Computer and Information Science",

publisher = "Springer Verlag",

pages = "419--431",

editor = "Isidoro Gitler and Jaime Klapp",

booktitle = "High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers",

address = "Alemania",

}

Rodríguez-Hernandez, A, Gómez-Torres, AM, del Valle-Gallegos, E, Jimenez-Escalante, J, Trost, N & Sanchez-Espinoza, VH 2016, Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU. in I Gitler & J Klapp (eds), High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers. Communications in Computer and Information Science, vol. 595, Springer Verlag, pp. 419-431, 6th International Conference on High Performance Computer Applications, ISUM 2015, Mexico City, Mexico, 9/03/15. https://doi.org/10.1007/978-3-319-32243-8_29

Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU. / Rodríguez-Hernandez, Andrés; Gómez-Torres, Armando M.; del Valle-Gallegos, Edmundo et al.
High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers. ed. / Isidoro Gitler; Jaime Klapp. Springer Verlag, 2016. p. 419-431 (Communications in Computer and Information Science; Vol. 595).

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

TY - GEN

T1 - Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU

AU - Rodríguez-Hernandez, Andrés

AU - Gómez-Torres, Armando M.

AU - del Valle-Gallegos, Edmundo

AU - Jimenez-Escalante, Javier

AU - Trost, Nico

AU - Sanchez-Espinoza, Victor H.

PY - 2016

Y1 - 2016

N2 - This paper presents the results of the accelerated solution of the linear algebraic system Av = b arising from a nodal finite element method implemented in the neutron diffusion code AZKIND to solve 3D problems. The numerical solution of full nuclear reactor cores with AZKIND implies the generation of large sparse algebraic systems that produce bottle-necks in the iterative solution. Aiming to alleviate the overload of the algorithm, an acceleration technique has to be implemented. Consequently, a Fortran plug-in of the open source linear algebra PARALUTION library (C ++) was integrated into the AZKIND source code (Fortran 95). This implementation allows AZKIND to use GPUs as well as CPUs, threading into the GPU thousands of arithmetic operations for parallel processing. Selected examples of preliminary investigations performed for a cluster of nuclear fuel assemblies are presented and the obtained results are discussed in this paper.

AB - This paper presents the results of the accelerated solution of the linear algebraic system Av = b arising from a nodal finite element method implemented in the neutron diffusion code AZKIND to solve 3D problems. The numerical solution of full nuclear reactor cores with AZKIND implies the generation of large sparse algebraic systems that produce bottle-necks in the iterative solution. Aiming to alleviate the overload of the algorithm, an acceleration technique has to be implemented. Consequently, a Fortran plug-in of the open source linear algebra PARALUTION library (C ++) was integrated into the AZKIND source code (Fortran 95). This implementation allows AZKIND to use GPUs as well as CPUs, threading into the GPU thousands of arithmetic operations for parallel processing. Selected examples of preliminary investigations performed for a cluster of nuclear fuel assemblies are presented and the obtained results are discussed in this paper.

UR - http://www.scopus.com/inward/record.url?scp=84964058378&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-32243-8_29

DO - 10.1007/978-3-319-32243-8_29

M3 - Contribución a la conferencia

SN - 9783319322421

T3 - Communications in Computer and Information Science

SP - 419

EP - 431

BT - High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers

A2 - Gitler, Isidoro

A2 - Klapp, Jaime

PB - Springer Verlag

T2 - 6th International Conference on High Performance Computer Applications, ISUM 2015

Y2 - 9 March 2015 through 13 March 2015

ER -

Rodríguez-Hernandez A, Gómez-Torres AM, del Valle-Gallegos E, Jimenez-Escalante J, Trost N, Sanchez-Espinoza VH. Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU. In Gitler I, Klapp J, editors, High Performance Computer Applications - 6th International Conference, ISUM 2015, Revised Selected Papers. Springer Verlag. 2016. p. 419-431. (Communications in Computer and Information Science). doi: 10.1007/978-3-319-32243-8_29

Accelerating AZKIND simulations of light water nuclear reactor cores using PARALUTION on GPU

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this