Use Cases of Computational Reproducibility for Scientific Workflows at Exascale

TL;DR

This paper introduces the ProvEn system to enhance reproducibility in scientific workflows by capturing provenance and performance data, demonstrated through climate simulations and molecular dynamics on HPC platforms.

Contribution

It presents a hybrid queriable system for capturing provenance and performance metrics, improving reproducibility in scientific workflows at exascale.

Findings

ProvEn effectively captures provenance and performance data.

Demonstrated improved reproducibility in climate and molecular dynamics workflows.

System supports querying and relating provenance and performance metrics.

Abstract

We propose an approach for improved reproducibility that includes capturing and relating provenance characteristics and performance metrics, in a hybrid queriable system, the ProvEn server. The system capabilities are illustrated on two use cases: scientific reproducibility of results in the ACME climate simulations and performance reproducibility in molecular dynamics workflows on HPC computing platforms.