Skip to content

performance.md

Latest commit

 

History

History
86 lines (58 loc) · 2.96 KB

performance.md

File metadata and controls

86 lines (58 loc) · 2.96 KB
Raw

Performance

FrankenNetworkX is built around the idea that speedups only matter if observable behavior stays compatible with NetworkX. The performance workflow therefore combines measurement, regression detection, and artifact retention.

What The Repo Measures

The checked-in benchmark pipeline currently tracks:

  • topology classes: grid, line, star, complete, erdos_renyi,
  • size buckets: small, medium, large,
  • latency tails (p95, p99) and memory (max_rss_kb),
  • hotspot backlog ranking and regression drift.

The main entry points are:

Artifacts land under artifacts/perf/.

Current Checked-In Gate Shape

The checked-in regression report is structured data that downstream tooling can consume directly:

import json
from pathlib import Path

report = json.loads(
    Path("artifacts/perf/phase2c/perf_regression_gate_report_v1.json").read_text()
)

assert report["status"] in {"pass", "fail"}
assert report["summary"]["scenario_count"] >= 1
assert "regression_count" in report["summary"]

SLO Surface

The SLO gate is configured in artifacts/perf/slo_thresholds.json.

It currently enforces or derives checks for:

  • shortest-path latency,
  • connected-components latency,
  • centrality latency,
  • flow latency,
  • I/O throughput,
  • mutation regression,
  • memory regression,
  • p99 tail regression.
import json
from pathlib import Path

thresholds = json.loads(Path("artifacts/perf/slo_thresholds.json").read_text())

assert len(thresholds["workloads"]) >= 5
assert len(thresholds["derived_checks"]) >= 3

Local Comparison Against NetworkX

Use the example benchmark script for a fast local sanity check:

python examples/benchmark_comparison.py

That script measures a small set of representative operations on identical graphs in NetworkX and FrankenNetworkX and prints JSON with median latency and speedup ratios.

CI Gate

The GitHub Actions G6 performance job runs the benchmark gate and uploads the resulting artifacts. The goal is not just to catch catastrophic regressions, but to keep a durable paper trail for why the performance posture changed.

Reading The Artifact Tree

  • artifacts/perf/phase2c/perf_baseline_matrix_v1.json stores the structured candidate matrix.
  • artifacts/perf/phase2c/perf_regression_gate_report_v1.json summarizes drift against the stored comparator.
  • artifacts/perf/cgse/ tracks deterministic tie-break policy benchmarks.
  • artifacts/perf/proof/ stores supporting evidence and proofs for specific optimizations.

For development workflow details, including rch usage for remote builds, see contributing.md.