Using Machine Learning at Scale in HPC Simulations with SmartSim: An Application to Ocean Climate Modeling
Preprint
Abstract
We demonstrate the first climate-scale, numerical ocean simulations improved through distributed, online inference of Deep Neural Networks (DNN) using SmartSim. SmartSim is a library dedicated to enabling online analysis and Machine Learning (ML) for traditional HPC simulations. In this paper, we detail the SmartSim architecture and provide benchmarks including online inference with a shared ML model on heterogeneous HPC systems. We demonstrate the capability of SmartSim by using it to run a 12-member ensemble of global-scale, high-resolution ocean simulations, each spanning 19 compute nodes, all communicating with the same ML architecture at each simulation timestep. In total, 970 billion inferences are collectively served by running the ensemble for a total of 120 simulated years. Finally, we show our solution is stable over the full duration of the model integrations, and that the inclusion of machine learning has minimal impact on the simulation runtimes.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- A turbulent eddy-viscosity surrogate modeling framework for Reynolds-averaged Navier-Stokes simulationsComputers & Fluids, 2021
- A Moist Physics Parameterization Based on Deep LearningJournal of Advances in Modeling Earth Systems, 2020
- The GFDL Global Ocean and Sea Ice Model OM4.0: Model Description and Simulation FeaturesJournal of Advances in Modeling Earth Systems, 2019
- Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networksJournal of Computational Physics, 2019
- Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme EventsJournal of Advances in Modeling Earth Systems, 2018
- Prognostic Validation of a Neural Network Unified Physics ParameterizationGeophysical Research Letters, 2018
- Earth System Modeling 2.0: A Blueprint for Models That Learn From Observations and Targeted High‐Resolution SimulationsGeophysical Research Letters, 2017
- Recent increase in oceanic carbon uptake driven by weaker upper-ocean overturningNature, 2017
- Energy budget-based backscatter in an eddy permitting primitive equation modelOcean Modelling, 2015
- Using a resolution function to regulate parameterizations of oceanic mesoscale eddy effectsOcean Modelling, 2013