Saved in:
Bibliographic Details
Main Authors: Fattebert, Jean-Luc, Negre, Christian F. A., Finkelstein, Joshua, Mohd-Yusof, Jamaludin, Osei-Kuffuor, Daniel, Wall, Michael E., Zhang, Yu, Bock, Nicolas, Mniszewski, Susan M.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.13772
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917574261866496
author Fattebert, Jean-Luc
Negre, Christian F. A.
Finkelstein, Joshua
Mohd-Yusof, Jamaludin
Osei-Kuffuor, Daniel
Wall, Michael E.
Zhang, Yu
Bock, Nicolas
Mniszewski, Susan M.
author_facet Fattebert, Jean-Luc
Negre, Christian F. A.
Finkelstein, Joshua
Mohd-Yusof, Jamaludin
Osei-Kuffuor, Daniel
Wall, Michael E.
Zhang, Yu
Bock, Nicolas
Mniszewski, Susan M.
contents To address the challenge of performance portability, and facilitate the implementation of electronic structure solvers, we developed the Basic Matrix Library (BML) and Parallel, Rapid O(N) and Graph-based Recursive Electronic Structure Solver (PROGRESS) libraries. BML implements linear algebra operations necessary for electronic structure kernels using a unified user interface for various matrix formats (dense, sparse) and architectures (CPUs, GPUs). Focusing on Density Functional Theory (DFT) and Tight-Binding (TB) models, PROGRESS implements several solvers for computing the single-particle density matrix and relies on BML. In this paper, we describe the general strategies used for these implementations on various computer architectures, using OpenMP target functionalities on GPUs, in conjunction with third-party libraries to handle performance critical numerical kernels. We demonstrate the portability of this approach and its performance on benchmark problems.
format Preprint
id arxiv_https___arxiv_org_abs_2401_13772
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Hybrid programming-model strategies for GPU offloading of electronic structure calculation kernels
Fattebert, Jean-Luc
Negre, Christian F. A.
Finkelstein, Joshua
Mohd-Yusof, Jamaludin
Osei-Kuffuor, Daniel
Wall, Michael E.
Zhang, Yu
Bock, Nicolas
Mniszewski, Susan M.
Computational Physics
Quantum Physics
To address the challenge of performance portability, and facilitate the implementation of electronic structure solvers, we developed the Basic Matrix Library (BML) and Parallel, Rapid O(N) and Graph-based Recursive Electronic Structure Solver (PROGRESS) libraries. BML implements linear algebra operations necessary for electronic structure kernels using a unified user interface for various matrix formats (dense, sparse) and architectures (CPUs, GPUs). Focusing on Density Functional Theory (DFT) and Tight-Binding (TB) models, PROGRESS implements several solvers for computing the single-particle density matrix and relies on BML. In this paper, we describe the general strategies used for these implementations on various computer architectures, using OpenMP target functionalities on GPUs, in conjunction with third-party libraries to handle performance critical numerical kernels. We demonstrate the portability of this approach and its performance on benchmark problems.
title Hybrid programming-model strategies for GPU offloading of electronic structure calculation kernels
topic Computational Physics
Quantum Physics
url https://arxiv.org/abs/2401.13772