Submitted Publication: WarpX and AMReX Performance Measurements and Science Runs

A data artifact for a publication submitted to Supercomputing 2022 includes performance measurements, source code patches, and larger science runs. The artifact is structured into separate archives for scripts, patched sources, and larger data outputs like profiles for GPU flop measurements. It was authored by Luca Fedeli and is available under an Open Access (green) license.

Use Cases

• Reproduce performance benchmarks based on the provided measurement scripts and setups.
• Analyze GPU flop performance profiles for computational kernels.
• Study the impact of source code patches on simulation performance.
• Validate scientific figures using the provided science run data.

Strengths

• Includes source code patches for the WarpX and AMReX frameworks, enabling reproducibility.
• Contains larger science runs that serve as source data for publication figures.
• Provides a central README file summarizing all measurements and experimental setups.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

Created as a data artifact for a submitted academic publication.