- Workload: Pipeline crosswalk with 40 design components, 100 SARIF findings, and 60 CVE records.
- Command: Inline harness running 30 iterations of
PipelineOrchestrator.runwith the probabilistic module disabled to capture the legacy baseline (see benchmark helper inperf/BENCHMARKS.csv). - Environment: Python 3.11 container, single worker process, no optional parsers installed.
- Runtime: ≈3.4057ms per run (average across 30 iterations).【8242a6†L1-L64】
- Peak Memory: 155.01KB traced via
tracemalloc(unchanged from earlier baseline).
This measurement captures the optimised crosswalk/token logic prior to enabling the probabilistic forecast engine so we can quantify its incremental overhead separately.