Saved in:
Bibliographic Details
Main Authors: Cortes, David, Juiz, Carlos, Bermejo, Belen
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.03263
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915477753692160
author Cortes, David
Juiz, Carlos
Bermejo, Belen
author_facet Cortes, David
Juiz, Carlos
Bermejo, Belen
contents Training large-scale deep learning models has become a key challenge for the scientific community and industry. While the massive use of GPUs can significantly speed up training times, this approach has a negative impact on efficiency. In this article, we present a detailed analysis of the times reported by MLPerf Training v4.1 on four workloads: BERT, Llama2 LoRA, RetinaNet, and Stable Diffusion, showing that there are configurations that optimise the relationship between performance, GPU usage, and efficiency. The results point to a break-even point that allows training times to be reduced while maximising efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2509_03263
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Estudio de la eficiencia en la escalabilidad de GPUs para el entrenamiento de Inteligencia Artificial
Cortes, David
Juiz, Carlos
Bermejo, Belen
Machine Learning
Artificial Intelligence
Performance
Training large-scale deep learning models has become a key challenge for the scientific community and industry. While the massive use of GPUs can significantly speed up training times, this approach has a negative impact on efficiency. In this article, we present a detailed analysis of the times reported by MLPerf Training v4.1 on four workloads: BERT, Llama2 LoRA, RetinaNet, and Stable Diffusion, showing that there are configurations that optimise the relationship between performance, GPU usage, and efficiency. The results point to a break-even point that allows training times to be reduced while maximising efficiency.
title Estudio de la eficiencia en la escalabilidad de GPUs para el entrenamiento de Inteligencia Artificial
topic Machine Learning
Artificial Intelligence
Performance
url https://arxiv.org/abs/2509.03263