Benchmarks

Benchmark Results

Recorded Criterion and stress/load results (local + Redis + Hybrid).

These numbers are not a controlled lab measurement. They were collected on a laptop with other services running, and Redis was run in Docker via compose.yaml.

If you want help interpreting the tables, read:

Benchmark Concepts

Reproducibility (this snapshot)

Field	Value
Repo	`https://github.com/dev-davexoyinbo/trypema`
Commit	`5fe93e7a3158c16e5143204798864797a2c4f689`
OS	macOS Tahoe 26.3
CPU	Apple M2 Pro
Memory	16GB
Rust	`rustc 1.93.0 (254b59607 2026-01-19)`
Docker	`29.2.0` (Docker Desktop `4.60.1`)
Docker Compose	`v5.0.2`
Redis	`redis:6.2-alpine` (via `compose.yaml`)
Redis URL	`redis://127.0.0.1:16379/` (default)
Hybrid sync interval	`sync_interval_ms = 10ms`
Notes	Non-controlled environment; results are directional

This section groups provider-specific comparisons into subpages:

What’s included

Criterion estimates: mean time per operation with a 95% confidence interval.
Units: ns for local benches, us for Redis benches.

Criterion HTML reports live under target/criterion/.

Open:

target/criterion/report/index.html

If your Criterion version outputs reports/ instead of report/, try:

target/criterion/reports/index.html

Benchmarking & Load Testing

Run Criterion microbenchmarks and the trypema-stress load harness reproducibly.

Local Benchmark Comparison

Local provider comparisons (stress/load + Criterion) on one host.