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

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.

References in corpus (18)

• Escaping free-energy minima
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models
• Ab initio theory and modeling of water
• MEAM potentials for Al, Si, Mg, Cu, and Fe alloys
• The Individual and Collective Effects of Exact Exchange and Dispersion Interactions on the Ab Initio Structure of Liquid Water
• Why does hydronium diffuse faster than hydroxide in liquid water?
• Deep neural network for the dielectric response of insulators
• Large scale and linear scaling DFT with the CONQUEST code
• Reinforced dynamics for enhanced sampling in large atomic and molecular systems
• Deep learning inter-atomic potential model for accurate irradiation damage simulations
• Isotope Effects in Liquid Water via Deep Potential Molecular Dynamics
• Linear scaling approach for atomistic calculation of excitonic properties of 10-million-atom nanostructures

Cited by in corpus (10)

• DeePMD-kit v2: A software package for Deep Potential models
• GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
• A deep potential model with long-range electrostatic interactions
• Condensed Matter Systems Exposed to Radiation: Multiscale Theory, Simulations, and Experiment
• Thermal Transport by Electrons and Ions in Warm Dense Aluminum: A Combined Density Functional Theory and Deep Potential Study
• Integrating Machine Learning with Physics-Based Modeling
• Ab Initio Generalized Langevin Equation
• Training Algorithm Matters for the Performance of Neural Network Potential: A Case Study of Adam and the Kalman Filter Optimizers
• Frame-independent vector-cloud neural network for nonlocal constitutive modeling on arbitrary grids
• Deep-learning interatomic potential for irradiation damage simulations in MoS2 with ab initial accuracy