Overcoming the Memory Bottleneck in Auxiliary Field Quantum Monte Carlo Simulations with Interpolative Separable Density Fitting

TL;DR

This paper introduces the use of interpolative separable density fitting (ISDF) to significantly reduce memory requirements in auxiliary field quantum Monte Carlo simulations, enabling more efficient computations for real materials.

Contribution

The paper demonstrates that ISDF can lower the memory scaling of AFQMC from quartic to quadratic in basis set size, improving computational efficiency.

Findings

Memory scaling reduced from O(M^4) to O(M^2)

Accurate structural properties of diamond carbon computed

Results compare favorably with existing methods and experiments

Abstract

We investigate the use of interpolative separable density fitting (ISDF) as a means to reduce the memory bottleneck in auxiliary field quantum Monte Carlo (AFQMC) simulations of real materials in Gaussian basis sets. We find that ISDF can reduce the memory scaling of AFQMC simulations from $\mathcal{O}(M^4)$ to $\mathcal{O}(M^2)$. We test these developments by computing the structural properties of Carbon in the diamond phase, comparing to results from existing computational methods and experiment.