lar-01 reproduction

Public source retained with the related article. Generated results, dependency directories, runtime storage, secrets, and oversized binary artifacts are intentionally excluded.

README

# LAR-01 measurement experiment

This synthetic Laravel 13 application produces distinct request, query, cache,
CPU, and queue signatures while preserving declared response behavior. It is
evidence for the LAR-01 article, not a benchmark of Brian's systems or a
portable capacity model.

The fixture uses SQLite and a registered delay function to make database work
observable on an isolated host. Absolute timings do not represent MySQL,
PostgreSQL, Redis, a network, or production hardware.

Run the complete profile in the pinned container:

```text
./run.sh
```

The runner writes raw samples and a derived summary under `results/`. Exact
latencies remain observations. Semantic verification uses response contracts
and broad directional signatures, not portable millisecond budgets.

After committing the complete fixture, reproduce it from a clean archive:

```text
./bin/run-clean-export.sh <commit>
```

Retained files