Quantum supercharger library: hyper-parallelism of the Hartree-Fock method

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Legacy electronic structure packages rely on standard Hartree-Fock (HF) computations.
• Existing algorithms can be computationally intensive, limiting the size and complexity of molecules studied.
• Hardware accelerators like Graphical Processing Units (GPUs) offer significant parallel processing capabilities.

Purpose of the Study:

• To develop and present a novel set of algorithms for massively parallel Hartree-Fock (HF) computations.
• To integrate these algorithms into a library named the Quantum Supercharger Library (QSL).
• To evaluate the performance enhancement of QSL utilizing GPU hardware.

Main Methods:

• Rewriting core Hartree-Fock (HF) algorithms for massive parallelization.
• Leveraging hardware accelerators, specifically Graphical Processing Units (GPUs).
• Optimizing one- and two-electron integral calculations and Self-Consistent Field (SCF) routines using GPU linear algebra libraries.

Main Results:

• The Quantum Supercharger Library (QSL) demonstrates significant acceleration of HF computations.
• QSL achieves up to a 20-fold speedup for HF 6-31G Self-Consistent Field (SCF) calculations on medium-sized molecules (e.g., buckyballs).
• Performance is compared against traditional Central Processing Unit (CPU) algorithms in established computational chemistry packages.

Conclusions:

• The developed algorithms and QSL effectively utilize GPU acceleration for quantum chemistry.
• Massive parallelization of HF computations offers a substantial performance improvement for electronic structure calculations.
• This approach enables faster and more efficient analysis of molecular systems in computational research.