Files
git.stella-ops.org/docs/features/checked/bench/benchmark-harness.md
2026-02-12 10:27:23 +02:00

2.4 KiB

Benchmark harness (reachability, scanner analyzers, policy engine, determinism)

Module

Bench

Status

VERIFIED

Description

Comprehensive benchmark harness code exists across LinkNotMerge, LinkNotMerge.Vex, Notify, PolicyEngine, and Scanner.Analyzers modules with deterministic benchmark/reporting support.

Implementation Details

  • LinkNotMerge Benchmark: src/Bench/StellaOps.Bench/LinkNotMerge/StellaOps.Bench.LinkNotMerge/ -- benchmark scenarios for linkset aggregation performance.
  • LinkNotMerge VEX Benchmark: src/Bench/StellaOps.Bench/LinkNotMerge.Vex/StellaOps.Bench.LinkNotMerge.Vex/ -- VEX-specific linkset benchmarks.
  • Notify Benchmark: src/Bench/StellaOps.Bench/Notify/StellaOps.Bench.Notify/ -- notification dispatch benchmarks.
  • PolicyEngine Benchmark: src/Bench/StellaOps.Bench/PolicyEngine/StellaOps.Bench.PolicyEngine/ -- policy evaluation benchmarks.
  • PolicyEngine Benchmark Policy: src/Bench/StellaOps.Bench/PolicyEngine/policies/benchmark-default.yaml -- benchmark policy fixture compatible with current StellaOps.Policy binder schema.
  • Scanner.Analyzers Benchmark: src/Bench/StellaOps.Bench/Scanner.Analyzers/StellaOps.Bench.ScannerAnalyzers/ -- scanner analyzer benchmarks.
  • Baseline Infrastructure: benchmark modules include Baseline/BaselineEntry.cs and Baseline/BaselineLoader.cs for ground-truth comparison.
  • Reporting Infrastructure: benchmark modules include JSON and Prometheus writers for machine-consumable artifacts.
  • Tests: link-not-merge, vex, notify, and scanner-analyzer benchmark test suites under src/Bench/StellaOps.Bench/**.Tests/.

E2E Test Plan

  • Run LinkNotMerge benchmark harness and verify scenario table output is generated.
  • Verify JSON report output is produced and non-empty.
  • Verify Prometheus metrics output is produced and non-empty.
  • Verify CSV result output is produced and non-empty.
  • Verify negative-path CLI behavior (--config missing path) exits non-zero.

Verification

  • Verified on 2026-02-11 via FLOW Tier 0/1/2 replay in run-005.
  • Tier 0: docs/qa/feature-checks/runs/bench/benchmark-harness/run-005/tier0-source-check.json
  • Tier 1: docs/qa/feature-checks/runs/bench/benchmark-harness/run-005/tier1-build-check.json
  • Tier 2: docs/qa/feature-checks/runs/bench/benchmark-harness/run-005/tier2-integration-check.json
  • Tier 2 evidence: docs/qa/feature-checks/runs/bench/benchmark-harness/run-005/evidence/