том 35 издание 4 страницы 312-324

GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP

Тип публикацииJournal Article
Дата публикации2021-04-19
scimago Q1
wos Q2
БС1
SJR1.051
CiteScore7.4
Impact factor2.5
ISSN10943420, 17412846
Hardware and Architecture
Software
Theoretical Computer Science
Краткое описание

Classical molecular dynamics (MD) calculations represent a significant part of the utilization time of high-performance computing systems. As usual, the efficiency of such calculations is based on an interplay of software and hardware that are nowadays moving to hybrid GPU-based technologies. Several well-developed open-source MD codes focused on GPUs differ both in their data management capabilities and in performance. In this work, we analyze the performance of LAMMPS, GROMACS and OpenMM MD packages with different GPU backends on Nvidia Volta and AMD Vega20 GPUs. We consider the efficiency of solving two identical MD models (generic for material science and biomolecular studies) using different software and hardware combinations. We describe our experience in porting the CUDA backend of LAMMPS to ROCm HIP that shows considerable benefits for AMD GPUs comparatively to the OpenCL backend.

Найдено 
Найдено 

Топ-30

Журналы

1
2
3
4
5
6
Journal of Molecular Liquids
6 публикаций, 7.79%
Lecture Notes in Computer Science
5 публикаций, 6.49%
Journal of Chemical Physics
4 публикации, 5.19%
Journal of Nuclear Materials
4 публикации, 5.19%
Communications in Computer and Information Science
3 публикации, 3.9%
Computer Physics Communications
3 публикации, 3.9%
Fluid Phase Equilibria
2 публикации, 2.6%
Energies
2 публикации, 2.6%
Journal of Parallel and Distributed Computing
2 публикации, 2.6%
Scientific Reports
1 публикация, 1.3%
Applied Physics Reviews
1 публикация, 1.3%
Membranes
1 публикация, 1.3%
TrAC - Trends in Analytical Chemistry
1 публикация, 1.3%
Computers in Biology and Medicine
1 публикация, 1.3%
Computational Materials Science
1 публикация, 1.3%
Journal of Physical Chemistry A
1 публикация, 1.3%
Journal of Experimental and Theoretical Physics
1 публикация, 1.3%
IEEE Access
1 публикация, 1.3%
Polymer Science - Series C
1 публикация, 1.3%
IET Science, Measurement and Technology
1 публикация, 1.3%
Physical Chemistry Chemical Physics
1 публикация, 1.3%
Astronomy and Computing
1 публикация, 1.3%
Smart Innovation, Systems and Technologies
1 публикация, 1.3%
International Journal of High Performance Computing Applications
1 публикация, 1.3%
International Journal of Molecular Sciences
1 публикация, 1.3%
High Temperature
1 публикация, 1.3%
Journal of Chemical Theory and Computation
1 публикация, 1.3%
Journal of Computer-Aided Molecular Design
1 публикация, 1.3%
Журнал Экспериментальной и Теоретической Физики
1 публикация, 1.3%
1
2
3
4
5
6

Издатели

5
10
15
20
25
Elsevier
22 публикации, 28.57%
Institute of Electrical and Electronics Engineers (IEEE)
13 публикаций, 16.88%
Springer Nature
11 публикаций, 14.29%
AIP Publishing
5 публикаций, 6.49%
MDPI
5 публикаций, 6.49%
Pleiades Publishing
4 публикации, 5.19%
Association for Computing Machinery (ACM)
4 публикации, 5.19%
American Chemical Society (ACS)
2 публикации, 2.6%
The Russian Academy of Sciences
2 публикации, 2.6%
Institution of Engineering and Technology (IET)
1 публикация, 1.3%
Royal Society of Chemistry (RSC)
1 публикация, 1.3%
SAGE
1 публикация, 1.3%
Cold Spring Harbor Laboratory
1 публикация, 1.3%
IGI Global
1 публикация, 1.3%
Oxford University Press
1 публикация, 1.3%
World Scientific
1 публикация, 1.3%
Srinivas University
1 публикация, 1.3%
5
10
15
20
25
  • Мы не учитываем публикации, у которых нет DOI.
  • Статистика публикаций обновляется еженедельно.

Вы ученый?

Создайте профиль, чтобы получать персональные рекомендации коллег, конференций и новых статей.
Метрики
77
Поделиться
Цитировать
ГОСТ |
Цитировать
Kondratyuk N. et al. GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP // International Journal of High Performance Computing Applications. 2021. Vol. 35. No. 4. pp. 312-324.
ГОСТ со всеми авторами (до 50) Скопировать
Kondratyuk N. D., Kondratyuk N., Nikolskiy V., Pavlov D., Stegailov V. GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP // International Journal of High Performance Computing Applications. 2021. Vol. 35. No. 4. pp. 312-324.
RIS |
Цитировать
TY - JOUR
DO - 10.1177/10943420211008288
UR - http://journals.sagepub.com/doi/10.1177/10943420211008288
TI - GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP
T2 - International Journal of High Performance Computing Applications
AU - Kondratyuk, Nikolay D.
AU - Kondratyuk, Nikolay
AU - Nikolskiy, Vsevolod
AU - Pavlov, Daniil
AU - Stegailov, Vladimir V.
PY - 2021
DA - 2021/04/19
PB - SAGE
SP - 312-324
IS - 4
VL - 35
SN - 1094-3420
SN - 1741-2846
ER -
BibTex |
Цитировать
BibTex (до 50 авторов) Скопировать
@article{2021_Kondratyuk,
author = {Nikolay D. Kondratyuk and Nikolay Kondratyuk and Vsevolod Nikolskiy and Daniil Pavlov and Vladimir V. Stegailov},
title = {GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP},
journal = {International Journal of High Performance Computing Applications},
year = {2021},
volume = {35},
publisher = {SAGE},
month = {apr},
url = {http://journals.sagepub.com/doi/10.1177/10943420211008288},
number = {4},
pages = {312--324},
doi = {10.1177/10943420211008288}
}
MLA
Цитировать
Kondratyuk, Nikolay, et al. “GPU-accelerated molecular dynamics: State-of-art software performance and porting from Nvidia CUDA to AMD HIP.” International Journal of High Performance Computing Applications, vol. 35, no. 4, Apr. 2021, pp. 312-324. http://journals.sagepub.com/doi/10.1177/10943420211008288.