Exploring Dynamic Load Balancing Algorithms for Block-Structured Mesh-and-Particle Simulations in AMReX

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Nanda, Amitash, Chowdhury, Md Kamal Hossain, Ross, Hannah, Gott, Kevin
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908467961266176
author Nanda, Amitash
Chowdhury, Md Kamal Hossain
Ross, Hannah
Gott, Kevin
author_facet Nanda, Amitash
Chowdhury, Md Kamal Hossain
Ross, Hannah
Gott, Kevin
contents Load balancing is critical for successful large-scale high-performance computing (HPC) simulations. With modern supercomputers increasing in complexity and variability, dynamic load balancing is becoming more critical to use computational resources efficiently. In this study, performed during a summer collaboration at Lawrence Berkeley National Laboratory, we investigate various standard dynamic load-balancing algorithms. This includes the time evaluation of a brute-force solve for application in algorithmic evaluation, as well as quality and time evaluations of the Knapsack algorithm, an SFC algorithm, and two novel algorithms: a painter's partition-based SFC algorithm and a combination Knapsack+SFC methodology-based on hardware topology. The results suggest Knapsack and painter's partition-based algorithms should be among the first algorithms evaluated by HPC codes for cases with limited weight deviation and will perform at least slightly better than AMReX's percentage-tracking partitioning strategy across most simulations, although effects diminish as weight variety increases.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15122
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring Dynamic Load Balancing Algorithms for Block-Structured Mesh-and-Particle Simulations in AMReX
Nanda, Amitash
Chowdhury, Md Kamal Hossain
Ross, Hannah
Gott, Kevin
Distributed, Parallel, and Cluster Computing
Load balancing is critical for successful large-scale high-performance computing (HPC) simulations. With modern supercomputers increasing in complexity and variability, dynamic load balancing is becoming more critical to use computational resources efficiently. In this study, performed during a summer collaboration at Lawrence Berkeley National Laboratory, we investigate various standard dynamic load-balancing algorithms. This includes the time evaluation of a brute-force solve for application in algorithmic evaluation, as well as quality and time evaluations of the Knapsack algorithm, an SFC algorithm, and two novel algorithms: a painter's partition-based SFC algorithm and a combination Knapsack+SFC methodology-based on hardware topology. The results suggest Knapsack and painter's partition-based algorithms should be among the first algorithms evaluated by HPC codes for cases with limited weight deviation and will perform at least slightly better than AMReX's percentage-tracking partitioning strategy across most simulations, although effects diminish as weight variety increases.
title Exploring Dynamic Load Balancing Algorithms for Block-Structured Mesh-and-Particle Simulations in AMReX
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2505.15122