Skip to content

Benchmarks

Recorded, reproducible benchmark reports for the packages whose implementation checklists require them. Each report is a standalone measurement you can re-run yourself; the scenario scripts live under benchmark/scenarios/.

What these measure - and what they don't

These are reference numbers for order-of-magnitude regression tracking, not controlled lab measurements. Two recurring results hold across every package:

  • Repository get() is O(1) - flat at ~0.08–0.11 µs across 10³–10⁵ populations, exactly as the design specifications require.
  • Throughput scenarios report honest sequential-vs-parallel numbers. For the trivial in-process fixtures, thread parallelism is GIL-bound and adds no speedup; real speedup materializes only when a concrete plugin backend (a native solver, an LLM call, a charting library) releases the GIL. The benchmarks prove contention-free correctness, not speedup.

Reports by package

  • Decision - RuleEngine throughput, DecisionAuditTrail latency.
  • Digital Twin - TwinRepository latency, cold-start replay cost.
  • Simulation - SimulationRunRepository latency, SimulationStateCache effectiveness.
  • Optimization - OptimizationRunRepository latency, sweep re-solve throughput.
  • AI Agents - TaskRepository latency, task-dispatch throughput.
  • Visualization - DashboardRepository latency, multi-widget render throughput.