Skip to main content
Work in Progress

This page is currently under development. Content may be incomplete or contain inaccuracies. If you notice any errors or have suggestions, please contact us.

Simulation & Modeling Metadata & Standards

Reproducing a simulation requires far more than just the output data — the exact software version, input parameters, mesh resolution, boundary conditions, and hardware used must all be documented. Metadata standards in this domain aim to capture this information systematically.

Metadata Standards and Frameworks​

MODA (Model Data)​

A semi-formal description framework for documenting simulation workflows, developed within the European Materials Modelling Council (EMMC).

  • Use: Documenting simulation workflows in materials modeling
  • Spec: EMMC MODA

W3C PROV (Provenance)​

A W3C standard for representing provenance information — who did what, when, and how. Can be applied to any computational workflow.

  • Use: Recording simulation lineage, parameter history, data transformations
  • Spec: W3C PROV

CodeMeta​

A metadata schema for research software, enabling proper citation and discovery of simulation codes.

  • Use: Documenting simulation software for citation and reproducibility
  • Spec: CodeMeta

CGNS Metadata​

The CFD General Notation System includes rich metadata support for mesh topology, boundary conditions, flow solutions, and convergence histories.

Essential Simulation Metadata​

Every simulation dataset should document:

Software Environment​

  • Solver name and version
  • Compiler and version (e.g., GCC 12.2, Intel oneAPI 2023)
  • MPI implementation and version
  • Operating system and HPC environment

Input Parameters​

  • Governing equations and models (e.g., RANS, LES, DNS for turbulence)
  • Material properties and constitutive models
  • Boundary and initial conditions
  • Time stepping scheme and step size

Mesh Information​

  • Element types and count
  • Mesh quality metrics (skewness, aspect ratio, orthogonality)
  • Refinement strategy (uniform, adaptive, local)

Computational Resources​

  • Number of cores/nodes used
  • Wall-clock time and CPU hours
  • Memory usage

Convergence and Validation​

  • Residual history
  • Grid convergence study results
  • Comparison with experimental or analytical reference data

Best Practices​

  1. Version-control input files — Use Git for solver input files, configuration, and post-processing scripts
  2. Automate metadata capture — Script the extraction of solver version, run parameters, and compute resources
  3. Store mesh and results together — Keep mesh, input, and output files in a coherent directory structure
  4. Document convergence — Include residual plots and grid independence studies with published results
  5. Use DVC or similar — For large binary files that don't fit in Git, use Data Version Control
  6. Create README files — Each simulation case directory should contain a README describing the setup