Korali: Efficient and Scalable Software Framework for Bayesian Uncertainty Quantification and Stochastic Optimization

TL;DR

Korali is an open-source software framework designed for large-scale Bayesian uncertainty quantification and stochastic optimization, efficiently utilizing massively-parallel architectures with fault tolerance and load balancing.

Contribution

It introduces a scalable, fault-tolerant, and efficient framework for Bayesian UQ and stochastic optimization, integrating with high-performance software and demonstrating superior performance.

Findings

Efficient scaling up to 512 nodes on supercomputers.

Outperforms existing software frameworks in benchmarks.

Successfully interfaces with multiphysics models like Aphros, Lammps, and Mirheo.

Abstract

We present Korali, an open-source framework for large-scale Bayesian uncertainty quantification and stochastic optimization. The framework relies on non-intrusive sampling of complex multiphysics models and enables their exploitation for optimization and decision-making. In addition, its distributed sampling engine makes efficient use of massively-parallel architectures while introducing novel fault tolerance and load balancing mechanisms. We demonstrate these features by interfacing Korali with existing high-performance software such as Aphros, Lammps (CPU-based), and Mirheo (GPU-based) and show efficient scaling for up to 512 nodes of the CSCS Piz Daint supercomputer. Finally, we present benchmarks demonstrating that Korali outperforms related state-of-the-art software frameworks.