Model Evaluation Service · View 07 of 21 · Structure
Decisions
- Control plane and execution plane scale independently: 3-4 pods against 0-600, on different node pools with different failure tolerance.
- The tool sandbox is its own component, not a library in the harness, because it executes model-generated calls.
Assumptions
- Peak 1,200 candidate generations per minute with a 4× burst for 30 minutes on release days (stated assumption).
- BigQuery as the score store because the dominant query is a per-slice comparison over tens of millions of rows.
Deliberate omission
- Edges are not drawn. Work flows through Pub/Sub and scores are read from BigQuery; the inference path is shown in the trust-zone view.