Skip to content

Measure throughput with subsecond precision#12

Closed
Toti330 wants to merge 1 commit into
mainfrom
feat/precise-timing
Closed

Measure throughput with subsecond precision#12
Toti330 wants to merge 1 commit into
mainfrom
feat/precise-timing

Conversation

@Toti330

@Toti330 Toti330 commented Jul 13, 2026

Copy link
Copy Markdown
Collaborator

Summary

Replaces whole-second throughput calculation with a precise floating-point elapsed duration.

Why this is useful

A 10,000-seed run that takes anywhere from 60.000 to 60.999 seconds previously displayed the same 166.66667 seeds/sec. This removes that quantization, so throughput comparisons can distinguish real changes from timing truncation.

Implementation

  • Uses Duration::as_secs_f64() for the final elapsed duration.
  • Prints elapsed seconds and seeds/sec with 6 decimal places.
  • Does not change worker, writer, channel, database, checkpoint, or scheduling behavior.

Independence

This PR is based directly on main and is intentionally standalone. It does not require any other performance or diagnostics PR.

Verification

  • cargo test --all-targets passed in the Rust Docker toolchain.
  • Benchmark smoke run printed 0.208606 elapsed seconds and 38.349896 seeds/sec, rather than an integer-second-derived value.

Generated by GPT-5.6 Terra & Sol. Ideas from the author; implementation by AI.

@SuperB3333

Copy link
Copy Markdown
Owner

Implemented in 83586fe

@SuperB3333 SuperB3333 closed this Jul 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants