```json { "table_name": "schema_migrations", "primary_key": "version", "candidates": [ { "rank": 0, "partition_key": "version", "partition_key_label": "primary_identifier", "partition_key_reasoning": "RPUV=1.00, uniqueness=1.000, perfect identifier for schema migration versions. Each version represents a unique database schema change. Semantically serves as the natural primary key for this dimension table.", "clustering_key": null, "clustering_key_label": null, "clustering_key_reasoning": "Dimension tables typically do not require clustering. The table is small (47 rows) and already organized by version number which has perfect monotonicity (1.000).", "world_line": null, "session_column": null, "confidence": "HIGH", "notes": "This is a classic schema migration tracking table (dimension type). The version column is the natural primary identifier with perfect uniqueness. The applied_at timestamp records when each migration was executed but does not serve as a partitioning axis. All metadata columns (_ingestion_timestamp, _batch_id, etc.) have RPUV=47 (single value) and are unsuitable for partitioning." } ] } ```