Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL

Keiichiro Fukazawa, Takeshi Soga, Takayuki Umeda, Takeshi Nanri

研究成果: 著書/レポートタイプへの貢献

抄録

The magnetohydrodynamic (MHD) simulation is often applied to study the global dynamics and configuration of a planetary magnetosphere for the space weather. In this paper, the computational performance of MHD code is evaluated with 128 nodes Xeon Phi KNL of Cray XC40. As the results, the 2D and 3D domain decompositions of SoA (structure of array) make the effective performances and AoS (array of structure) and hybrid parallel computation become low performances. Adding the performance optimizations for Xeon Phi to our MHD simulation code, then we have obtained 2.4 % increase of execution efficiency in total and we achieved 3 TFlops performance gain using 128 nodes.

元の言語英語
ホスト出版物のタイトルParallel Computing is Everywhere
編集者Gerhard R. Joubert, Patrizio Dazzi, Frans Peters, Marco Danelutto, Sanzio Bassini
出版者IOS Press BV
ページ178-187
ページ数10
ISBN(電子版)9781614998426
DOI
出版物ステータス出版済み - 1 1 2018

出版物シリーズ

名前Advances in Parallel Computing
32
ISSN(印刷物)0927-5452
ISSN(電子版)1879-808X

Fingerprint

Magnetosphere
Magnetohydrodynamics
Decomposition

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

これを引用

Fukazawa, K., Soga, T., Umeda, T., & Nanri, T. (2018). Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL. : G. R. Joubert, P. Dazzi, F. Peters, M. Danelutto, & S. Bassini (版), Parallel Computing is Everywhere (pp. 178-187). (Advances in Parallel Computing; 巻数 32). IOS Press BV. https://doi.org/10.3233/978-1-61499-843-3-178

Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL. / Fukazawa, Keiichiro; Soga, Takeshi; Umeda, Takayuki; Nanri, Takeshi.

Parallel Computing is Everywhere. 版 / Gerhard R. Joubert; Patrizio Dazzi; Frans Peters; Marco Danelutto; Sanzio Bassini. IOS Press BV, 2018. p. 178-187 (Advances in Parallel Computing; 巻 32).

研究成果: 著書/レポートタイプへの貢献

Fukazawa, K, Soga, T, Umeda, T & Nanri, T 2018, Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL. : GR Joubert, P Dazzi, F Peters, M Danelutto & S Bassini (版), Parallel Computing is Everywhere. Advances in Parallel Computing, 巻. 32, IOS Press BV, pp. 178-187. https://doi.org/10.3233/978-1-61499-843-3-178
Fukazawa K, Soga T, Umeda T, Nanri T. Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL. : Joubert GR, Dazzi P, Peters F, Danelutto M, Bassini S, 編集者, Parallel Computing is Everywhere. IOS Press BV. 2018. p. 178-187. (Advances in Parallel Computing). https://doi.org/10.3233/978-1-61499-843-3-178
Fukazawa, Keiichiro ; Soga, Takeshi ; Umeda, Takayuki ; Nanri, Takeshi. / Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL. Parallel Computing is Everywhere. 編集者 / Gerhard R. Joubert ; Patrizio Dazzi ; Frans Peters ; Marco Danelutto ; Sanzio Bassini. IOS Press BV, 2018. pp. 178-187 (Advances in Parallel Computing).
@inbook{dd0ccbcda1a940d09a0a66b5d9aceac8,
title = "Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL",
abstract = "The magnetohydrodynamic (MHD) simulation is often applied to study the global dynamics and configuration of a planetary magnetosphere for the space weather. In this paper, the computational performance of MHD code is evaluated with 128 nodes Xeon Phi KNL of Cray XC40. As the results, the 2D and 3D domain decompositions of SoA (structure of array) make the effective performances and AoS (array of structure) and hybrid parallel computation become low performances. Adding the performance optimizations for Xeon Phi to our MHD simulation code, then we have obtained 2.4 {\%} increase of execution efficiency in total and we achieved 3 TFlops performance gain using 128 nodes.",
author = "Keiichiro Fukazawa and Takeshi Soga and Takayuki Umeda and Takeshi Nanri",
year = "2018",
month = "1",
day = "1",
doi = "10.3233/978-1-61499-843-3-178",
language = "English",
series = "Advances in Parallel Computing",
publisher = "IOS Press BV",
pages = "178--187",
editor = "Joubert, {Gerhard R.} and Patrizio Dazzi and Frans Peters and Marco Danelutto and Sanzio Bassini",
booktitle = "Parallel Computing is Everywhere",

}

TY - CHAP

T1 - Performance Evaluation and Optimization of MagnetoHydroDynamic Simulation for Planetary Magnetosphere with Xeon Phi KNL

AU - Fukazawa, Keiichiro

AU - Soga, Takeshi

AU - Umeda, Takayuki

AU - Nanri, Takeshi

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The magnetohydrodynamic (MHD) simulation is often applied to study the global dynamics and configuration of a planetary magnetosphere for the space weather. In this paper, the computational performance of MHD code is evaluated with 128 nodes Xeon Phi KNL of Cray XC40. As the results, the 2D and 3D domain decompositions of SoA (structure of array) make the effective performances and AoS (array of structure) and hybrid parallel computation become low performances. Adding the performance optimizations for Xeon Phi to our MHD simulation code, then we have obtained 2.4 % increase of execution efficiency in total and we achieved 3 TFlops performance gain using 128 nodes.

AB - The magnetohydrodynamic (MHD) simulation is often applied to study the global dynamics and configuration of a planetary magnetosphere for the space weather. In this paper, the computational performance of MHD code is evaluated with 128 nodes Xeon Phi KNL of Cray XC40. As the results, the 2D and 3D domain decompositions of SoA (structure of array) make the effective performances and AoS (array of structure) and hybrid parallel computation become low performances. Adding the performance optimizations for Xeon Phi to our MHD simulation code, then we have obtained 2.4 % increase of execution efficiency in total and we achieved 3 TFlops performance gain using 128 nodes.

UR - http://www.scopus.com/inward/record.url?scp=85043570970&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043570970&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-843-3-178

DO - 10.3233/978-1-61499-843-3-178

M3 - Chapter

AN - SCOPUS:85043570970

T3 - Advances in Parallel Computing

SP - 178

EP - 187

BT - Parallel Computing is Everywhere

A2 - Joubert, Gerhard R.

A2 - Dazzi, Patrizio

A2 - Peters, Frans

A2 - Danelutto, Marco

A2 - Bassini, Sanzio

PB - IOS Press BV

ER -