Porting WarpX to GPU-accelerated platforms

A. Myers; A. Almgren; L. D. Amorim; J. Bell; L. Fedeli; L. Ge; K. Gott; D. P. Grote; M. Hogan; A. Huebl; R. Jambunathan; R. Lehe; C. Ng; M. Rowan; O. Shapoval; M. Thévenet; J. L. Vay; H. Vincenti; E. Yang; N. Zaïm; W. Zhang; Y. Zhao; E. Zoni

doi:10.1016/j.parco.2021.102833

Porting WarpX to GPU-accelerated platforms

A. Myers^*
, A. Almgren
, L. D. Amorim
, J. Bell
, L. Fedeli
, L. Ge
, K. Gott
, D. P. Grote
, M. Hogan
, A. Huebl
, R. Jambunathan
, R. Lehe
, C. Ng
, M. Rowan
, O. Shapoval
, M. Thévenet
, J. L. Vay
, H. Vincenti
, E. Yang
, N. Zaïm

W. Zhang, Y. Zhao, E. Zoni

^*Corresponding author for this work

Lawrence Berkeley National Laboratory
Université Paris-Saclay
SLAC National Accelerator Laboratory
Lawrence Livermore National Laboratory
German Electron Synchrotron

Research output: Contribution to journal › Article › peer-review

43 Scopus citations

Abstract

WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures. We describe the strategy, based on the AMReX library, followed to allow WarpX to use the GPU-accelerated nodes on OLCF's Summit supercomputer, a strategy we believe will extend to the upcoming machines Frontier and Aurora. We summarize the challenges encountered, lessons learned, and give current performance results on a series of relevant benchmark problems.

Original language	English
Article number	102833
Journal	Parallel Computing
Volume	108
DOIs	https://doi.org/10.1016/j.parco.2021.102833
State	Published - Dec 2021
Externally published	Yes

Keywords

Accelerator modeling
Exascale Computing
Particle-in-cell methods

Access to Document

10.1016/j.parco.2021.102833

Cite this

Myers, A., Almgren, A., Amorim, L. D., Bell, J., Fedeli, L., Ge, L., Gott, K., Grote, D. P., Hogan, M., Huebl, A., Jambunathan, R., Lehe, R., Ng, C., Rowan, M., Shapoval, O., Thévenet, M., Vay, J. L., Vincenti, H., Yang, E., ... Zoni, E. (2021). Porting WarpX to GPU-accelerated platforms. Parallel Computing, 108, Article 102833. https://doi.org/10.1016/j.parco.2021.102833

@article{6eb34908fe5a4a0eb6f5222eba06a1f9,

title = "Porting WarpX to GPU-accelerated platforms",

abstract = "WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures. We describe the strategy, based on the AMReX library, followed to allow WarpX to use the GPU-accelerated nodes on OLCF's Summit supercomputer, a strategy we believe will extend to the upcoming machines Frontier and Aurora. We summarize the challenges encountered, lessons learned, and give current performance results on a series of relevant benchmark problems.",

keywords = "Accelerator modeling, Exascale Computing, Particle-in-cell methods",

author = "A. Myers and A. Almgren and Amorim, \{L. D.\} and J. Bell and L. Fedeli and L. Ge and K. Gott and Grote, \{D. P.\} and M. Hogan and A. Huebl and R. Jambunathan and R. Lehe and C. Ng and M. Rowan and O. Shapoval and M. Th{\'e}venet and Vay, \{J. L.\} and H. Vincenti and E. Yang and N. Za{\"i}m and W. Zhang and Y. Zhao and E. Zoni",

note = "Publisher Copyright: {\textcopyright} 2021",

year = "2021",

month = dec,

doi = "10.1016/j.parco.2021.102833",

language = "英语",

volume = "108",

journal = "Parallel Computing",

issn = "0167-8191",

publisher = "Elsevier B.V.",

}

Myers, A, Almgren, A, Amorim, LD, Bell, J, Fedeli, L, Ge, L, Gott, K, Grote, DP, Hogan, M, Huebl, A, Jambunathan, R, Lehe, R, Ng, C, Rowan, M, Shapoval, O, Thévenet, M, Vay, JL, Vincenti, H, Yang, E, Zaïm, N, Zhang, W, Zhao, Y & Zoni, E 2021, 'Porting WarpX to GPU-accelerated platforms', Parallel Computing, vol. 108, 102833. https://doi.org/10.1016/j.parco.2021.102833

TY - JOUR

T1 - Porting WarpX to GPU-accelerated platforms

AU - Myers, A.

AU - Almgren, A.

AU - Amorim, L. D.

AU - Bell, J.

AU - Fedeli, L.

AU - Ge, L.

AU - Gott, K.

AU - Grote, D. P.

AU - Hogan, M.

AU - Huebl, A.

AU - Jambunathan, R.

AU - Lehe, R.

AU - Ng, C.

AU - Rowan, M.

AU - Shapoval, O.

AU - Thévenet, M.

AU - Vay, J. L.

AU - Vincenti, H.

AU - Yang, E.

AU - Zaïm, N.

AU - Zhang, W.

AU - Zhao, Y.

AU - Zoni, E.

PY - 2021/12

Y1 - 2021/12

N2 - WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures. We describe the strategy, based on the AMReX library, followed to allow WarpX to use the GPU-accelerated nodes on OLCF's Summit supercomputer, a strategy we believe will extend to the upcoming machines Frontier and Aurora. We summarize the challenges encountered, lessons learned, and give current performance results on a series of relevant benchmark problems.

AB - WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures. We describe the strategy, based on the AMReX library, followed to allow WarpX to use the GPU-accelerated nodes on OLCF's Summit supercomputer, a strategy we believe will extend to the upcoming machines Frontier and Aurora. We summarize the challenges encountered, lessons learned, and give current performance results on a series of relevant benchmark problems.

KW - Accelerator modeling

KW - Exascale Computing

KW - Particle-in-cell methods

UR - https://www.scopus.com/pages/publications/85115656406

U2 - 10.1016/j.parco.2021.102833

DO - 10.1016/j.parco.2021.102833

M3 - 文章

AN - SCOPUS:85115656406

SN - 0167-8191

VL - 108

JO - Parallel Computing

JF - Parallel Computing

M1 - 102833

ER -

Porting WarpX to GPU-accelerated platforms

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this