Database Properties
Database Type: PostgreSQL - 18.4 (Ubuntu 18.4-1.pgdg24.04+1)
Schema public
standard public schema
Tables
| Table / View | Children | Parents | Columns | Type | Comments |
|---|---|---|---|---|---|
| scrape_bench_steps | 0 | 1 | 7 | Table | {“what”:“One step in a scraping benchmark run, identified by run_id and measured_at.”,“why”:“Without it, per-step performance and status history would be lost.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:02:04.649161178+00:00”} |
| bench | 0 | 0 | 20 | Table | {“what”:“One row per LLM inference request in a benchmark run.”,“why”:“Without it, per-request performance and cost data would be lost.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:01:55.711478241+00:00”} |
| scrape_bench_runs | 1 | 0 | 10 | Table | {“what”:“One row per benchmark run of the scraping pipeline, capturing environment, versions, and outcome.”,“why”:“Without it, there is no way to correlate timing, config, and results across runs.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:02:04.543586423+00:00”} |
| bench_cache | 0 | 0 | 13 | Table | {“what”:“Cache of per-turn benchmark results keyed by run, provider, model, arm, and turn.”,“why”:“Without it, repeated benchmark queries would recompute costs and token counts.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T04:04:15.146918278+00:00”} |
| bench_card | 0 | 0 | 23 | Table | {“what”:“A benchmark card with metadata and content.”,“why”:“Without it, benchmark cases cannot be stored or retrieved.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T04:04:48.041962044+00:00”} |
| bench_latency | 0 | 0 | 15 | Table | {“what”:“One row per latency measurement from a benchmark run.”,“why”:“Without it, benchmark performance cannot be evaluated or compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:02:00.818767390+00:00”} |
| bench_concurrency | 0 | 0 | 19 | Table | {“what”:“One row per concurrency test level within a benchmark run.”,“why”:“Groups per-level metrics so results can be compared across runs and levels.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:01:51.827746339+00:00”} |
| bench_agentic | 0 | 0 | 19 | Table | {“what”:“One row per agentic benchmark run; identifies a single trial.”,“why”:“Without it, results from different runs and models cannot be compared or aggregated.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T04:03:57.485927147+00:00”} |
| bench_run | 0 | 0 | 22 | Table | {“what”:“A single benchmark test execution with its configuration.”,“why”:“Without it, runs cannot be uniquely identified or compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T04:04:47.689322246+00:00”} |