workspace-scale-benchmark

name	workspace-scale-benchmark
description	Run and interpret Grove workspace scale performance benchmarks using `benchmark-scale`, including baseline capture, regression comparison, and optimization loops. Use when user asks about performance, scale, regressions, startup/discovery speed, polling/update throughput, p50/p95 timings, or benchmark harness usage. Trigger phrases: "benchmark-scale", "workspace scale", "perf regression", "p95", "performance baseline", "issue 38".
allowed-tools	Read, Grep, Glob, Task

Workspace Scale Benchmark

Use this skill to measure and improve Grove performance at workspace scale.

Objective

Produce repeatable performance measurements for:

• discovery
• status-target-generation
• sort-update-pipeline

Then use those measurements to guide and verify optimizations.

Commands

# Human-readable benchmark report (N=10,100,500)
cargo run -- benchmark-scale

# JSON report for machine parsing
cargo run -- benchmark-scale --json

# Create or refresh baseline JSON
cargo run -- benchmark-scale --write-baseline docs/workspace-scale-baseline.json

# Compare current run vs baseline (informational warnings)
cargo run -- benchmark-scale --baseline docs/workspace-scale-baseline.json

# Tighten/loosen warning threshold (default 35)
cargo run -- benchmark-scale --baseline docs/workspace-scale-baseline.json --warn-regression-pct 25

Workflow

1. Confirm baseline exists.

• If missing, create one with --write-baseline.

2. Collect current benchmark report.

• Run once for quick signal.
• Run 3 times for noisy investigations.

3. Identify dominant flow by p95.

• Prioritize the largest p95 first.

4. Map flow to code area and form one optimization hypothesis.

• discovery:
  • src/infrastructure/adapters/*.rs
  • src/application/services/discovery_service.rs
• status-target-generation:
  • src/application/agent_runtime/polling.rs
• sort-update-pipeline:
  • src/application/agent_runtime/reconciliation.rs
  • status/update paths called from polling pipeline

5. Implement one focused change.

• Keep scope small, avoid mixed refactors.

6. Re-run benchmark against baseline.

• Use --baseline.
• If warning appears once but not consistently, treat as noise and rerun.

7. Validate quality gates.

• Run make precommit.
• Run tests you changed.

8. Report before/after p50 and p95 for affected flow(s).

Interpretation Rules

• Use p95 for regression decisions.
• Use p50 to gauge typical user experience.
• Single-run spikes are not enough, confirm with repeated runs.
• Baseline warnings are informational only, not hard failures.

Guardrails

• Do not overwrite baseline unless asked, or when intentionally resetting perf targets.
• If baseline changes intentionally, update both:
  • docs/workspace-scale-baseline.json
  • docs/workspace-scale-benchmarks.md
• Keep environment notes when sharing numbers (OS, Rust version, run date).

Handoff Checklist

• Commands run
• Baseline file used
• Flows improved or regressed
• Before/after p50+p95 values
• Any uncertainty due to run-to-run noise