Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics

This work evaluates the computing performance of finite element analysis in structural mechanics using modern multi-GPU systems. We can avoid the usual memory limitations when using one GPU device for many-core computing using multiple GPUs for scientific computing. We use a GPU-awareness MPI approa...

Full description

Saved in:
Bibliographic Details
Main Authors: David Herrero-Pérez, Humberto Martínez-Barberá
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/3/1095
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850199494324387840
author David Herrero-Pérez
Humberto Martínez-Barberá
author_facet David Herrero-Pérez
Humberto Martínez-Barberá
author_sort David Herrero-Pérez
collection DOAJ
description This work evaluates the computing performance of finite element analysis in structural mechanics using modern multi-GPU systems. We can avoid the usual memory limitations when using one GPU device for many-core computing using multiple GPUs for scientific computing. We use a GPU-awareness MPI approach implementing a suitable smoothed aggregation multigrid for preconditioning an iterative distributed conjugate gradient solver for GPU computing. We evaluate the performance and scalability of different models, problem sizes, and computing resources. We take an efficient multi-core implementation as the reference to assess the computing performance of the numerical results. The numerical results show the advantages and limitations of using distributed many-core architectures to address structural mechanics problems.
format Article
id doaj-art-4434925cb6d747b6a87a88f6b6c613e2
institution OA Journals
issn 2076-3417
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-4434925cb6d747b6a87a88f6b6c613e22025-08-20T02:12:37ZengMDPI AGApplied Sciences2076-34172025-01-01153109510.3390/app15031095Multi-GPU Acceleration for Finite Element Analysis in Structural MechanicsDavid Herrero-Pérez0Humberto Martínez-Barberá1Escuela Técnica Superior de Ingeniería Industrial, Universidad Politécnica de Cartagena, Campus Muralla del Mar, 30202 Cartagena, SpainFacultad de Informática, Universidad de Murcia, 30100 Murcia, SpainThis work evaluates the computing performance of finite element analysis in structural mechanics using modern multi-GPU systems. We can avoid the usual memory limitations when using one GPU device for many-core computing using multiple GPUs for scientific computing. We use a GPU-awareness MPI approach implementing a suitable smoothed aggregation multigrid for preconditioning an iterative distributed conjugate gradient solver for GPU computing. We evaluate the performance and scalability of different models, problem sizes, and computing resources. We take an efficient multi-core implementation as the reference to assess the computing performance of the numerical results. The numerical results show the advantages and limitations of using distributed many-core architectures to address structural mechanics problems.https://www.mdpi.com/2076-3417/15/3/1095GPU computingmulti-GPU systemsstructural mechanicsalgebraic multigrid
spellingShingle David Herrero-Pérez
Humberto Martínez-Barberá
Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
Applied Sciences
GPU computing
multi-GPU systems
structural mechanics
algebraic multigrid
title Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
title_full Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
title_fullStr Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
title_full_unstemmed Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
title_short Multi-GPU Acceleration for Finite Element Analysis in Structural Mechanics
title_sort multi gpu acceleration for finite element analysis in structural mechanics
topic GPU computing
multi-GPU systems
structural mechanics
algebraic multigrid
url https://www.mdpi.com/2076-3417/15/3/1095
work_keys_str_mv AT davidherreroperez multigpuaccelerationforfiniteelementanalysisinstructuralmechanics
AT humbertomartinezbarbera multigpuaccelerationforfiniteelementanalysisinstructuralmechanics