In Situ Framework for Coupling Simulation and Machine Learning with Application to CFD
This work addresses the problem of efficient ML-simulation coupling for computational fluid dynamics researchers, offering an incremental improvement by streamlining existing workflows.
The authors tackled the I/O and storage bottlenecks in coupling machine learning with large-scale simulations by developing an in situ framework that simplifies integration and enables training and inference workflows on heterogeneous clusters, demonstrating perfect scaling efficiency on the Polaris supercomputer with negligible overhead relative to solver time steps.
Recent years have seen many successful applications of machine learning (ML) to facilitate fluid dynamic computations. As simulations grow, generating new training datasets for traditional offline learning creates I/O and storage bottlenecks. Additionally, performing inference at runtime requires non-trivial coupling of ML framework libraries with simulation codes. This work offers a solution to both limitations by simplifying this coupling and enabling in situ training and inference workflows on heterogeneous clusters. Leveraging SmartSim, the presented framework deploys a database to store data and ML models in memory, thus circumventing the file system. On the Polaris supercomputer, we demonstrate perfect scaling efficiency to the full machine size of the data transfer and inference costs thanks to a novel co-located deployment of the database. Moreover, we train an autoencoder in situ from a turbulent flow simulation, showing that the framework overhead is negligible relative to a solver time step and training epoch.