Pushing the limit of molecular dynamics with ab initio accuracy to 100 million atoms with machine learning

TL;DR

This paper demonstrates that machine learning-based molecular dynamics can simulate over 100 million atoms with ab initio accuracy, achieving unprecedented scale and speed on supercomputers, thus enabling new scientific possibilities.

Contribution

The authors introduce a GPU-accelerated Deep Potential Molecular Dynamics protocol capable of simulating 100 million atoms with ab initio accuracy at high efficiency and scalability.

Findings

Simulates over 1 nanosecond trajectories of 100 million atoms per day.

Achieves 91 PFLOPS in double precision on Summit supercomputer.

Scales efficiently up to the entire Summit supercomputer.

Abstract

For 35 years, {\it ab initio} molecular dynamics (AIMD) has been the method of choice for modeling complex atomistic phenomena from first principles. However, most AIMD applications are limited by computational cost to systems with thousands of atoms at most. We report that a machine learning-based simulation protocol (Deep Potential Molecular Dynamics), while retaining {\it ab initio} accuracy, can simulate more than 1 nanosecond-long trajectory of over 100 million atoms per day, using a highly optimized code (GPU DeePMD-kit) on the Summit supercomputer. Our code can efficiently scale up to the entire Summit supercomputer, attaining $91$ PFLOPS in double precision ($45.5\%$ of the peak) and {$162$/$275$ PFLOPS in mixed-single/half precision}. The great accomplishment of this work is that it opens the door to simulating unprecedented size and time scales with {\it ab initio} accuracy. It also poses new challenges to the next-generation supercomputer for a better integration of machine learning and physical modeling.