Enregistré dans:
Détails bibliographiques
Auteurs principaux: Tonarelli, Melanie, Riva, Simone, Benedusi, Pietro, Ferrandi, Fabrizio, Krause, Rolf
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2511.01573
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918183642857472
author Tonarelli, Melanie
Riva, Simone
Benedusi, Pietro
Ferrandi, Fabrizio
Krause, Rolf
author_facet Tonarelli, Melanie
Riva, Simone
Benedusi, Pietro
Ferrandi, Fabrizio
Krause, Rolf
contents We introduce a distributed adaptive quadrature method that formulates multidimensional integration as a hierarchical domain decomposition problem on multi-GPU architectures. The integration domain is recursively partitioned into subdomains whose refinement is guided by local error estimators. Each subdomain evolves independently on a GPU, which exposes a significant load imbalance as the adaptive process progresses. To address this challenge, we introduce a decentralised load redistribution schemes based on a cyclic round-robin policy. This strategy dynamically rebalance subdomains across devices through non-blocking, CUDA-aware MPI communication that overlaps with computation. The proposed strategy has two main advantages compared to a state-of-the-art GPU-tailored package: higher efficiency in high dimensions; and improved robustness w.r.t the integrand regularity and the target accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2511_01573
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Multidimensional Quadrature on Multi-GPU Systems
Tonarelli, Melanie
Riva, Simone
Benedusi, Pietro
Ferrandi, Fabrizio
Krause, Rolf
Distributed, Parallel, and Cluster Computing
We introduce a distributed adaptive quadrature method that formulates multidimensional integration as a hierarchical domain decomposition problem on multi-GPU architectures. The integration domain is recursively partitioned into subdomains whose refinement is guided by local error estimators. Each subdomain evolves independently on a GPU, which exposes a significant load imbalance as the adaptive process progresses. To address this challenge, we introduce a decentralised load redistribution schemes based on a cyclic round-robin policy. This strategy dynamically rebalance subdomains across devices through non-blocking, CUDA-aware MPI communication that overlaps with computation. The proposed strategy has two main advantages compared to a state-of-the-art GPU-tailored package: higher efficiency in high dimensions; and improved robustness w.r.t the integrand regularity and the target accuracy.
title Adaptive Multidimensional Quadrature on Multi-GPU Systems
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2511.01573