Towards Electronic Structure-Based Ab-Initio Molecular Dynamics Simulations with Hundreds of Millions of Atoms

TL;DR

This paper demonstrates the ability to perform electronic structure-based ab-initio molecular dynamics simulations on unprecedented scales exceeding 100 million atoms by leveraging novel linear-scaling methods, GPU acceleration, and error compensation techniques.

Contribution

The authors introduce the NOLSM method, enabling scalable, high-performance AIMD simulations on massively parallel GPU systems at an unprecedented scale.

Findings

Achieved simulation of over 100 million atoms.

Reached a performance of 324 PFLOP/s on 1536 GPUs.

Demonstrated efficient scaling and accuracy with error compensation.

Abstract

We push the boundaries of electronic structure-based \textit{ab-initio} molecular dynamics (AIMD) beyond 100 million atoms. This scale is otherwise barely reachable with classical force-field methods or novel neural network and machine learning potentials. We achieve this breakthrough by combining innovations in linear-scaling AIMD, efficient and approximate sparse linear algebra, low and mixed-precision floating-point computation on GPUs, and a compensation scheme for the errors introduced by numerical approximations. The core of our work is the non-orthogonalized local submatrix method (NOLSM), which scales very favorably to massively parallel computing systems and translates large sparse matrix operations into highly parallel, dense matrix operations that are ideally suited to hardware accelerators. We demonstrate that the NOLSM method, which is at the center point of each AIMD step, is able to achieve a sustained performance of 324 PFLOP/s in mixed FP16/FP32 precision corresponding to an efficiency of 67.7% when running on 1536 NVIDIA A100 GPUs.