Accelerating Quantum Many-body Calculations in and out of Equilibrium

Quantum many-body methods are essential for understanding excited-state phenomena in and out of equilibrium. While these approaches provide high accuracy in describing the properties of complex materials, their applicability is often limited by steep computational costs. In this talk, we present strategies to accelerate quantum many-body calculations by leveraging emerging hardware capabilities and software engineering techniques. We will demonstrate how GW calculations of systems of thousands to over 10,000 atoms can be realized by adapting to exascale GPU architecture [1]. In addition, we will show how non-equilibrium time-dependent many-body calculations can be accelerated via high-performance computing programming schemes. Taken together, we present efficient workflows to enable the simulation of excited-state dynamics for complex materials at the many-body level.



[1] B. Zhang, D. Weinberg, C.-E. Hsu, A. R Altman, Y. Shi, J. B. White III, D. Vigil-Fowler, S. G. Louie, J. R. Deslippe, F. H. da Jornada, Z. Li, M. Del Ben, Advancing Quantum Many-Body GW Calculations on Exascale Supercomputing Platforms, arXiv:2509.23018 (2025).