Metadata practices for simulation workflows

TL;DR

This paper proposes flexible, software-agnostic metadata practices for simulation workflows, including a Python tool called Archivist, to improve reproducibility, sharing, and understanding of complex simulation data across scientific disciplines.

Contribution

It introduces general, adaptable practices for acquiring and structuring simulation metadata, along with the Archivist tool, to enhance reproducibility and data management in diverse scientific workflows.

Findings

Practices are applicable across different simulation domains.

Archivist effectively structures metadata for high-performance computing workflows.

Enhances reproducibility and data sharing in simulation research.

Abstract

Computer simulations are an essential pillar of knowledge generation in science. Exploring, understanding, reproducing, and sharing the results of simulations relies on tracking and organizing the metadata describing the numerical experiments. The models used to understand real-world systems, and the computational machinery required to simulate them, are typically complex, and produce large amounts of heterogeneous metadata. Here, we present general practices for acquiring and handling metadata that are agnostic to software and hardware, and highly flexible for the user. These consist of two steps: 1) recording and storing raw metadata, and 2) selecting and structuring metadata. As a proof of concept, we develop the Archivist, a Python tool to help with the second step, and use it to apply our practices to distinct high-performance computing use cases from neuroscience and hydrology. Our practices and the Archivist can readily be applied to existing workflows without the need for substantial restructuring. They support sustainable numerical workflows, fostering replicability, reproducibility, data exploration, and data sharing in simulation-based research.