perf(ci): reduce GitHub Actions costs ~60-65% across all workflows
Analysis of Feb 17 data showed 400+ workflow runs/day consuming ~398 billable minutes (~200 hours/month projected). Implemented targeted optimizations: High-impact changes: - sec-audit.yml: add path filters (Cargo.toml, src/**, crates/**, deny.toml); skip docs-only PRs - test-benchmarks.yml: move from every-push-to-main to weekly schedule; retention 30d -> 7d - pub-docker-img.yml: tighten PR smoke build path filters to Docker-specific files only - sec-codeql.yml: reduce from twice-daily (14 runs/week) to weekly Medium-impact changes: - ci-run.yml: merge lint + lint-strict-delta into single job; drop --release from smoke build - feature-matrix.yml: remove push trigger (weekly-only); remove redundant cargo test step - dependabot.yml: monthly instead of weekly; reduce PR limits from 11 to 5/month; group all deps Runner cost savings: - Switch 6 lightweight API-only workflows to ubuntu-latest (PR Labeler, Intake, Auto Responder, Check Stale, Check Status, Sync Contributors) - pr-check-status.yml: reduce from every 12h to daily New files: - docs/ci-cost-optimization.md: comprehensive analysis and revised architecture documentation - scripts/ci/fetch_actions_data.py: reusable GitHub Actions cost analysis script Estimated impact: daily billable minutes ~400 -> ~120-150 (60-65%% reduction), monthly hours ~200 -> ~60-75, Dependabot PRs ~44/month -> ~5 (89%% reduction)
This commit is contained in:
Alex Gorevski
parent
8f7d879fd5
commit
44725da08c
15 changed files with 512 additions and 85 deletions
156
scripts/ci/fetch_actions_data.py
Normal file
156
scripts/ci/fetch_actions_data.py
Normal file
|
|
@ -0,0 +1,156 @@
|
|||
#!/usr/bin/env python3
|
||||
"""Fetch GitHub Actions workflow runs for a given date and summarize costs."""
|
||||
|
||||
import json
|
||||
import subprocess
|
||||
import sys
|
||||
from datetime import datetime, timezone
|
||||
|
||||
|
||||
def fetch_runs(repo, date_str, page=1, per_page=100):
|
||||
"""Fetch completed workflow runs for a given date."""
|
||||
url = (
|
||||
f"https://api.github.com/repos/{repo}/actions/runs"
|
||||
f"?created={date_str}&per_page={per_page}&page={page}"
|
||||
)
|
||||
result = subprocess.run(
|
||||
["curl", "-sS", "-H", "Accept: application/vnd.github+json", url],
|
||||
capture_output=True, text=True
|
||||
)
|
||||
return json.loads(result.stdout)
|
||||
|
||||
|
||||
def fetch_jobs(repo, run_id):
|
||||
"""Fetch jobs for a specific run."""
|
||||
url = f"https://api.github.com/repos/{repo}/actions/runs/{run_id}/jobs?per_page=100"
|
||||
result = subprocess.run(
|
||||
["curl", "-sS", "-H", "Accept: application/vnd.github+json", url],
|
||||
capture_output=True, text=True
|
||||
)
|
||||
return json.loads(result.stdout)
|
||||
|
||||
|
||||
def parse_duration(started, completed):
|
||||
"""Return duration in seconds between two ISO timestamps."""
|
||||
if not started or not completed:
|
||||
return 0
|
||||
try:
|
||||
s = datetime.fromisoformat(started.replace("Z", "+00:00"))
|
||||
c = datetime.fromisoformat(completed.replace("Z", "+00:00"))
|
||||
return max(0, (c - s).total_seconds())
|
||||
except Exception:
|
||||
return 0
|
||||
|
||||
|
||||
def main():
|
||||
repo = "zeroclaw-labs/zeroclaw"
|
||||
date_str = "2026-02-17"
|
||||
|
||||
print(f"Fetching workflow runs for {repo} on {date_str}...")
|
||||
print("=" * 100)
|
||||
|
||||
all_runs = []
|
||||
for page in range(1, 5): # up to 400 runs
|
||||
data = fetch_runs(repo, date_str, page=page)
|
||||
runs = data.get("workflow_runs", [])
|
||||
if not runs:
|
||||
break
|
||||
all_runs.extend(runs)
|
||||
if len(runs) < 100:
|
||||
break
|
||||
|
||||
print(f"Total workflow runs found: {len(all_runs)}")
|
||||
print()
|
||||
|
||||
# Group by workflow name
|
||||
workflow_stats = {}
|
||||
for run in all_runs:
|
||||
name = run.get("name", "Unknown")
|
||||
event = run.get("event", "unknown")
|
||||
conclusion = run.get("conclusion", "unknown")
|
||||
run_id = run.get("id")
|
||||
|
||||
if name not in workflow_stats:
|
||||
workflow_stats[name] = {
|
||||
"count": 0,
|
||||
"events": {},
|
||||
"conclusions": {},
|
||||
"total_job_seconds": 0,
|
||||
"total_jobs": 0,
|
||||
"run_ids": [],
|
||||
}
|
||||
|
||||
workflow_stats[name]["count"] += 1
|
||||
workflow_stats[name]["events"][event] = workflow_stats[name]["events"].get(event, 0) + 1
|
||||
workflow_stats[name]["conclusions"][conclusion] = workflow_stats[name]["conclusions"].get(conclusion, 0) + 1
|
||||
workflow_stats[name]["run_ids"].append(run_id)
|
||||
|
||||
# For each workflow, sample up to 3 runs to get job-level timing
|
||||
print("Sampling job-level timing (up to 3 runs per workflow)...")
|
||||
print()
|
||||
|
||||
for name, stats in workflow_stats.items():
|
||||
sample_ids = stats["run_ids"][:3]
|
||||
for run_id in sample_ids:
|
||||
jobs_data = fetch_jobs(repo, run_id)
|
||||
jobs = jobs_data.get("jobs", [])
|
||||
for job in jobs:
|
||||
started = job.get("started_at")
|
||||
completed = job.get("completed_at")
|
||||
duration = parse_duration(started, completed)
|
||||
stats["total_job_seconds"] += duration
|
||||
stats["total_jobs"] += 1
|
||||
|
||||
# Extrapolate: if we sampled N runs but there are M total, scale up
|
||||
sampled = len(sample_ids)
|
||||
total = stats["count"]
|
||||
if sampled > 0 and sampled < total:
|
||||
scale = total / sampled
|
||||
stats["estimated_total_seconds"] = stats["total_job_seconds"] * scale
|
||||
else:
|
||||
stats["estimated_total_seconds"] = stats["total_job_seconds"]
|
||||
|
||||
# Print summary sorted by estimated cost (descending)
|
||||
sorted_workflows = sorted(
|
||||
workflow_stats.items(),
|
||||
key=lambda x: x[1]["estimated_total_seconds"],
|
||||
reverse=True
|
||||
)
|
||||
|
||||
print("=" * 100)
|
||||
print(f"{'Workflow':<40} {'Runs':>5} {'SampledJobs':>12} {'SampledMins':>12} {'Est.TotalMins':>14} {'Events'}")
|
||||
print("-" * 100)
|
||||
|
||||
grand_total_minutes = 0
|
||||
for name, stats in sorted_workflows:
|
||||
sampled_mins = stats["total_job_seconds"] / 60
|
||||
est_total_mins = stats["estimated_total_seconds"] / 60
|
||||
grand_total_minutes += est_total_mins
|
||||
events_str = ", ".join(f"{k}={v}" for k, v in stats["events"].items())
|
||||
conclusions_str = ", ".join(f"{k}={v}" for k, v in stats["conclusions"].items())
|
||||
print(
|
||||
f"{name:<40} {stats['count']:>5} {stats['total_jobs']:>12} "
|
||||
f"{sampled_mins:>12.1f} {est_total_mins:>14.1f} {events_str}"
|
||||
)
|
||||
print(f"{'':>40} {'':>5} {'':>12} {'':>12} {'':>14} outcomes: {conclusions_str}")
|
||||
|
||||
print("-" * 100)
|
||||
print(f"{'GRAND TOTAL':>40} {len(all_runs):>5} {'':>12} {'':>12} {grand_total_minutes:>14.1f}")
|
||||
print(f"\nEstimated total billable minutes on {date_str}: {grand_total_minutes:.0f} min ({grand_total_minutes/60:.1f} hours)")
|
||||
print()
|
||||
|
||||
# Also show raw run list
|
||||
print("\n" + "=" * 100)
|
||||
print("DETAILED RUN LIST")
|
||||
print("=" * 100)
|
||||
for run in all_runs:
|
||||
name = run.get("name", "Unknown")
|
||||
event = run.get("event", "unknown")
|
||||
conclusion = run.get("conclusion", "unknown")
|
||||
run_id = run.get("id")
|
||||
started = run.get("run_started_at", "?")
|
||||
print(f" [{run_id}] {name:<40} conclusion={conclusion:<12} event={event:<20} started={started}")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Loading…
Add table
Add a link
Reference in a new issue