Distributed Equivariant Graph Neural Networks for Large-Scale Electronic Structure Prediction
OPEN ACCESS
Loading...
Author / Producer
Date
2025-07-04
Publication Type
Working Paper
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
Equivariant Graph Neural Networks (eGNNs) trained on density-functional theory (DFT) data can potentially perform electronic structure prediction at unprecedented scales, enabling investigation of the electronic properties of materials with extended defects, interfaces, or exhibiting disordered phases. However, as interactions between atomic orbitals typically extend over 10+ angstroms, the graph representations required for this task tend to be densely connected, and the memory requirements to perform training and inference on these large structures can exceed the limits of modern GPUs. Here we present a distributed eGNN implementation which leverages direct GPU communication and introduce a partitioning strategy of the input graph to reduce the number of embedding exchanges between GPUs. Our implementation shows strong scaling up to 128 GPUs, and weak scaling up to 512 GPUs with 87% parallel efficiency for structures with 3,000 to 190,000 atoms on the Alps supercomputer.
Permanent link
Publication status
published
Editor
Book title
Journal / series
Volume
Pages / Article No.
Publisher
Cornell University
Event
Edition / version
v1
Methods
Software
Geographic location
Date collected
Date created
Subject
Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Distributed, Parallel, and Cluster Computing (cs.DC); Computational Physics (physics.comp-ph)
Organisational unit
03925 - Luisier, Mathieu / Luisier, Mathieu
Notes
Funding
198612 - Advanced Learning Methods On Dedicated nano-Devices (ALMOND) (SNF)