Benchmarks
The maintained benchmarks show ergo-sbe is not slower than official sbe-tool
on equal work. They are not a speed claim. Method, fairness rules, ceilings
and the gate:
sbe/BENCHMARKS.md.
| Question | Answer |
|---|---|
| Run it | just bench (SBE), just bench-cluster (cluster) |
| What is graded | ergon / sbe-tool, Criterion’s mean, highest ratio over 5 repeats, both LTO profiles |
| Pass | Highest ratio ≤ ceiling (1.00 for most pairs) |
| Inconclusive (also passes) | Above the ceiling, but its confidence range overlaps the A/A control’s noise band |
| Fail | Above the ceiling and outside the A/A band, or no A/A band measured |
| Tolerance | none (zero tolerance): no slack is added to a ceiling |
| Where results go | target/bench-runs/<run-id>/ (just bench); target/criterion and target/bench-no-lto/criterion (just bench-cluster, overwritten) |
This book records no benchmark result. Quote a ratio with its run id, commit, host, rustc and profile, or do not quote it.
Build your own code with lto = true and codegen-units = 1. Generated
hot-path methods carry #[inline], so a non-LTO build can inline them too.
Also compare against your own previous release: see Methodology.