Benchmark Results

This page stores raw benchmark snapshots (machine + date + tool versions + command outputs).

If you want interpretation or tool-choice advice, see Performance Comparisons.

How to Read This Page

• Treat each section as a point-in-time snapshot, not a universal ranking.
• Compare runs only when dataset, commands, and versions are aligned.
• Prefer ratios (tool A vs tool B on same host) over absolute wall-clock numbers.

Snapshot Format (Use This for New Entries)

When adding a new run, append a section with:

1. Date (UTC)
2. Git commit
3. Hardware + OS
4. Tool versions
5. Scenario table (copy directly from benchmarks/comparison_results/*.md)
6. Notes (anything unusual: thermal throttling, background load, etc.)

Minimal template

## System: <machine-name> (<YYYY-MM-DD>)

- Commit: `<sha>`
- CPU: <model>
- Cores/threads: <value>
- RAM: <value>
- OS: <value>
- Tool versions: <list>

### Results

| Scenario | Tool | Time | Throughput | Notes |
|---|---|---:|---:|---|
| ... | ... | ... | ... | ... |

### Notes

- ...

Current Snapshots

Apple M1 (2024-10-26)

• Commit: development build at measurement time
• OS: macOS (Darwin)
• CPU/RAM: Apple M1, 8 cores (4P+4E), 16 GB
• Tools: jq 1.7.1, ripgrep 14.1.0, BSD awk, miller 6.12.0, qsv 0.131.0

Source artifacts: benchmarks/comparison_results/01_simple_filter.md through 07_csv.md (from that run).

Intel Core i5 (2025-10-26)

• Commit: 09faf43
• OS: macOS 15.6.1 (x86_64)
• CPU/RAM: Intel Core i5 @ 3.0 GHz, 6 cores, 16 GB
• Tools: grep 2.6.0, ripgrep 15.1.0, jq 1.6, mlr 6.15.0, qsv 8.1.1, angle-grinder 0.19.5, klp 0.77.0

Source artifacts: benchmarks/comparison_results/01_simple_filter.md through 07_csv.md (from that run).

Add or Refresh a Snapshot

cargo build --release
./benchmarks/generate_comparison_data.sh
just bench-compare

Then copy the generated markdown tables from benchmarks/comparison_results/ into a new section using the template above.