In-Transit Data Transport Strategies for Coupled AI-Simulation Workflow Patterns

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tummalapalli, Harikrishna, Balin, Riccardo, Simpson, Christine M., Park, Andrew, Alsaadi, Aymen, Shao, Andrew E., Brewer, Wesley, Jha, Shantenu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916964764483584
author Tummalapalli, Harikrishna
Balin, Riccardo
Simpson, Christine M.
Park, Andrew
Alsaadi, Aymen
Shao, Andrew E.
Brewer, Wesley
Jha, Shantenu
author_facet Tummalapalli, Harikrishna
Balin, Riccardo
Simpson, Christine M.
Park, Andrew
Alsaadi, Aymen
Shao, Andrew E.
Brewer, Wesley
Jha, Shantenu
contents Coupled AI-Simulation workflows are becoming the major workloads for HPC facilities, and their increasing complexity necessitates new tools for performance analysis and prototyping of new in-situ workflows. We present SimAI-Bench, a tool designed to both prototype and evaluate these coupled workflows. In this paper, we use SimAI-Bench to benchmark the data transport performance of two common patterns on the Aurora supercomputer: a one-to-one workflow with co-located simulation and AI training instances, and a many-to-one workflow where a single AI model is trained from an ensemble of simulations. For the one-to-one pattern, our analysis shows that node-local and DragonHPC data staging strategies provide excellent performance compared Redis and Lustre file system. For the many-to-one pattern, we find that data transport becomes a dominant bottleneck as the ensemble size grows. Our evaluation reveals that file system is the optimal solution among the tested strategies for the many-to-one pattern.
format Preprint
id arxiv_https___arxiv_org_abs_2509_19150
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle In-Transit Data Transport Strategies for Coupled AI-Simulation Workflow Patterns
Tummalapalli, Harikrishna
Balin, Riccardo
Simpson, Christine M.
Park, Andrew
Alsaadi, Aymen
Shao, Andrew E.
Brewer, Wesley
Jha, Shantenu
Distributed, Parallel, and Cluster Computing
Coupled AI-Simulation workflows are becoming the major workloads for HPC facilities, and their increasing complexity necessitates new tools for performance analysis and prototyping of new in-situ workflows. We present SimAI-Bench, a tool designed to both prototype and evaluate these coupled workflows. In this paper, we use SimAI-Bench to benchmark the data transport performance of two common patterns on the Aurora supercomputer: a one-to-one workflow with co-located simulation and AI training instances, and a many-to-one workflow where a single AI model is trained from an ensemble of simulations. For the one-to-one pattern, our analysis shows that node-local and DragonHPC data staging strategies provide excellent performance compared Redis and Lustre file system. For the many-to-one pattern, we find that data transport becomes a dominant bottleneck as the ensemble size grows. Our evaluation reveals that file system is the optimal solution among the tested strategies for the many-to-one pattern.
title In-Transit Data Transport Strategies for Coupled AI-Simulation Workflow Patterns
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2509.19150