Integrating Deep Learning in Domain Sciences at Exascale

Archibald, Rick; Chow, Edmond; D'Azevedo, Eduardo F.; Dongarra, Jack; Eisenbach, Markus; Febbo, Rocco; Lopez, Florent; Nichols, Daniel; Tomov, Stanimire; Wong, Kwai; Yin, Junqi

doi:10.1007/978-3-030-63393-6_3

Cited by 3 publications

(1 citation statement)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial intelligence (AI) methods have become pervasive in recent years due to numerous algorithmic advances and the accessibility of computational power (Archibald et al, 2020; Jouppi et al, 2018). In computational fluid dynamics (CFD), AI has been used to assist, accelerate, enhance, or even replace existing physical solvers based on partial differential equations (PDEs) (Brunton et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

A study on the performance of distributed training of data-driven CFD simulations

Iserte

González-Barberá

Barreda

et al. 2023

The International Journal of High Performance Computing Applica

View full text Add to dashboard Cite

Data-driven methods for computer simulations are blooming in many scientific areas. The traditional approach to simulating physical behaviors relies on solving partial differential equations (PDEs). Since calculating these iterative equations is highly both computationally demanding and time-consuming, data-driven methods leverage artificial intelligence (AI) techniques to alleviate that workload. Data-driven methods have to be trained in advance to provide their subsequent fast predictions; however, the cost of the training stage is non-negligible. This article presents a predictive model for inferencing future states of a specific fluid simulation that serves as a use case for evaluating different training alternatives. Particularly, this study compares the performance of only CPU, multi-GPU, and distributed approaches for training a time series forecasting deep learning model. With some slight code adaptations, results show and compare, in different implementations, the benefits of distributed GPU-enabled training for predicting high-accuracy states in a fraction of the time needed by the computational fluid dynamics solver.

show abstract

Section: Introductionmentioning

confidence: 99%