High-speed and low-power molecular dynamics processing unit (MDPU) with ab initio accuracy

Abstract Molecular dynamics (MD) is an indispensable atomistic-scale computational tool widely-used in various disciplines. In the past decades, nearly all ab initio MD and machine-learning MD have been based on the general-purpose central/graphics processing units (CPU/GPU), which are well-known to suffer from their intrinsic “memory wall” and “power wall” bottlenecks. Consequently, nowadays MD calculations with ab initio accuracy are extremely time-consuming and power-consuming, imposing serious restrictions on the MD simulation size and duration. To solve this problem, here we propose a special-purpose MD processing unit (MDPU), which could reduce MD time and power consumption by about 10 3 times (10 9 times) compared to state-of-the-art machine-learning MD (ab initio MD) based on CPU/GPU, while keeping ab initio accuracy. With significantly-enhanced performance, the proposed MDPU may pave a way for the accurate atomistic-scale analysis of large-size and/or long-duration problems which were impossible/impractical to compute before.

Authors

Institutions

Publication Details

Journal
npj Computational Materials
Published
2024-11-07
DOI
https://doi.org/10.1038/s41524-024-01422-3
Citations
8
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.94

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

High-speed and low-power molecular dynamics processing unit (MDPU) with ab initio accuracy

Denghui Lu, Jun Cheng, Hanhan Sun, Yujia Zhang et al.
8 citations
npj Computational Materials
Machine Learning in Materials Science
0.94
article

High-speed and low-power molecular dynamics processing unit (MDPU) with ab initio accuracy

Denghui Lu, Jun Cheng, Hanhan Sun, Yujia Zhang, Bin Liang, Xi Ding, Pinghui Mo, Xin Zhang, Zhuoying Zhao, E Weinan, Mohan Chen, Jianxing Huang, Jiayu Dai, Bo Chen, Junhua Li, Fei Liu, Xinyu Li, Yinan Wang, Handong Wang, Jie Liu, Duo Zhang, Linfeng Zhang, Dawei Guan, Mengchao Shi
article en
8 citations

Abstract

Abstract Molecular dynamics (MD) is an indispensable atomistic-scale computational tool widely-used in various disciplines. In the past decades, nearly all ab initio MD and machine-learning MD have been based on the general-purpose central/graphics processing units (CPU/GPU), which are well-known to suffer from their intrinsic “memory wall” and “power wall” bottlenecks. Consequently, nowadays MD calculations with ab initio accuracy are extremely time-consuming and power-consuming, imposing serious restrictions on the MD simulation size and duration. To solve this problem, here we propose a special-purpose MD processing unit (MDPU), which could reduce MD time and power consumption by about 10 3 times (10 9 times) compared to state-of-the-art machine-learning MD (ab initio MD) based on CPU/GPU, while keeping ab initio accuracy. With significantly-enhanced performance, the proposed MDPU may pave a way for the accurate atomistic-scale analysis of large-size and/or long-duration problems which were impossible/impractical to compute before.

npj Computational MaterialsVol. 10(1)
Hunan University (CN), National University of Defense Technology (CN), Xiamen University (CN), University of Washington (US), Peking University (CN), Bay Institute (US), Institute of Applied Physics and Computational Mathematics (CN), Collaborative Innovation Center of Chemistry for Energy Materials (CN)
National Natural Science Foundation of China
Affordable and clean energy
Openalex Percentile: Top 23%
Machine Learning in Materials Science
0.94
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.