GPU-accelerated dynamic nonlinear optimization with ExaModels and MadNLP

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pacaud, François, Shin, Sungho
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912023578673152
author Pacaud, François
Shin, Sungho
author_facet Pacaud, François
Shin, Sungho
contents We investigate the potential of Graphics Processing Units (GPUs) to solve large-scale nonlinear programs with a dynamic structure. Using ExaModels, a GPU-accelerated automatic differentiation tool, and the interior-point solver MadNLP, we significantly reduce the time to solve dynamic nonlinear optimization problems. The sparse linear systems formulated in the interior-point method is solved on the GPU using a hybrid solver combining an iterative method with a sparse Cholesky factorization, which harness the newly released NVIDIA cuDSS solver. Our results on the classical distillation column instance show that despite a significant pre-processing time, the hybrid solver allows to reduce the time per iteration by a factor of 25 for the largest instance.
format Preprint
id arxiv_https___arxiv_org_abs_2403_15913
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GPU-accelerated dynamic nonlinear optimization with ExaModels and MadNLP
Pacaud, François
Shin, Sungho
Optimization and Control
We investigate the potential of Graphics Processing Units (GPUs) to solve large-scale nonlinear programs with a dynamic structure. Using ExaModels, a GPU-accelerated automatic differentiation tool, and the interior-point solver MadNLP, we significantly reduce the time to solve dynamic nonlinear optimization problems. The sparse linear systems formulated in the interior-point method is solved on the GPU using a hybrid solver combining an iterative method with a sparse Cholesky factorization, which harness the newly released NVIDIA cuDSS solver. Our results on the classical distillation column instance show that despite a significant pre-processing time, the hybrid solver allows to reduce the time per iteration by a factor of 25 for the largest instance.
title GPU-accelerated dynamic nonlinear optimization with ExaModels and MadNLP
topic Optimization and Control
url https://arxiv.org/abs/2403.15913