English

Advanced Techniques for High-Performance Fock Matrix Construction on GPU Clusters

Computational Physics 2024-08-01 v1 Materials Science Chemical Physics Quantum Physics

Abstract

This Article presents two optimized multi-GPU algorithms for Fock matrix construction, building on the work of Ufimtsev et al. and Barca et al. The novel algorithms, opt-UM and opt-Brc, introduce significant enhancements, including improved integral screening, exploitation of sparsity and symmetry, a linear scaling exchange matrix assembly algorithm, and extended capabilities for Hartree-Fock caculations up to ff-type angular momentum functions. Opt-Brc excels for smaller systems and for highly contracted triple-ζ\zeta basis sets, while opt-UM is advantageous for large molecular systems. Performance benchmarks on NVIDIA A100 GPUs show that our algorithms in the EXtreme-scale Electronic Structure System (EXESS), when combined, outperform all current GPU and CPU Fock build implementations in TeraChem, QUICK, GPU4PySCF, LibIntX, ORCA, and Q-Chem. The implementations were benchmarked on linear and globular systems and average speed ups across three double-ζ\zeta basis sets of 1.5×\times, 5.2×\times, and 8.5×\times were observed compared to TeraChem, GPU4PySCF, and QUICK respectively. Strong scaling analysis reveals over 91% parallel efficiency on four GPUs for opt-Brc, making it typically faster for multi-GPU execution. Single-compute-node comparisons with CPU-based software like ORCA and Q-Chem show speedups of up to 42×\times and 31×\times, respectively, enhancing power efficiency by up to 18×\times.

Keywords

Cite

@article{arxiv.2407.21445,
  title  = {Advanced Techniques for High-Performance Fock Matrix Construction on GPU Clusters},
  author = {Elise Palethorpe and Ryan Stocks and Giuseppe M. J. Barca},
  journal= {arXiv preprint arXiv:2407.21445},
  year   = {2024}
}
R2 v1 2026-06-28T17:59:05.774Z