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.
| Original language | English |
|---|---|
| Title of host publication | Parallel Computing is Everywhere |
| Editors | Gerhard R. Joubert, Patrizio Dazzi, Frans Peters, Marco Danelutto, Sanzio Bassini |
| Publisher | IOS Press BV |
| Pages | 178-187 |
| Number of pages | 10 |
| ISBN (Electronic) | 9781614998426 |
| DOIs | |
| Publication status | Published - Jan 1 2018 |
Publication series
| Name | Advances in Parallel Computing |
|---|---|
| Volume | 32 |
| ISSN (Print) | 0927-5452 |
| ISSN (Electronic) | 1879-808X |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Science(all)
Cite this
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. ed. / Gerhard R. Joubert; Patrizio Dazzi; Frans Peters; Marco Danelutto; Sanzio Bassini. IOS Press BV, 2018. p. 178-187 (Advances in Parallel Computing; Vol. 32).Research output: Chapter in Book/Report/Conference proceeding › Chapter
}
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 -