These files are committed simulator outputs from the default benchmark sweep.
Generated with:
python scripts/run_benchmarks.py --output-dir artifacts/default_sweep
python scripts/plot_results.py --results artifacts/default_sweep/results.json --out-dir artifacts/default_sweep/plotsFiles:
results.json: full benchmark recordsresults.csv: flattened table for quick filteringrun_metadata.json: exact command plus scenario, policy, and workload listsplots/: policy-level summary figures
These artifacts are simulator results on synthetic traces. They are intended for inspection and reproduction, not as claims about real runtime throughput.