```json { "table_name": "analysis_run_llm_usage", "primary_key": "jobs_id,model_name", "candidates": [ { "rank": 0, "partition_key": "jobs_id", "partition_key_label": "primary_identifier", "partition_key_reasoning": "RPUV=2.00 indicates each job has exactly 2 rows (one per model type). This is the primary entity organizing LLM usage records. Top 3 coverage is only 2.65%, showing good distribution across 113 unique jobs. Semantically represents the analysis run whose LLM usage is being tracked.", "clustering_key": null, "clustering_key_label": null, "clustering_key_reasoning": "Dimension tables typically do not require clustering. The model_name is part of the composite key but not a clustering axis.", "world_line": null, "session_column": null, "confidence": "HIGH", "notes": "This is a dimension table tracking LLM usage per analysis job. The composite key (jobs_id, model_name) uniquely identifies each row, as each job uses multiple models. The RPUV=2.00 pattern suggests a consistent two-model usage pattern per job." } ] } ```