Where should chunked text be written before it is indexed?
Why D is the answer
Delta tables in Unity Catalog
Delta tables in Unity Catalog keep chunks governed, and a vector index can build from them directly.
In detail
- Chunked text is written into Delta Lake tables in Unity Catalog.
- That gives you governed data that an index can build from.
- Keep the source document ID on each row, for citations.
Use case
An illustrative scenario, not a real incident.
A team writes chunks to a Delta table in Unity Catalog. The chunks are now governed and versioned, and an index can build directly from them.
See it in different ways
flowchart LR A["Source documents"] --> B["Extract text"] B --> C["Filter extraneous content"] C --> D["Chunk"] D --> E["Delta table in Unity Catalog"] E --> F["Vector Search index"]
stateDiagram-v2 [*] --> Raw Raw --> Text: parse or OCR Text --> Clean: remove extraneous content Clean --> Chunked: split Chunked --> Governed: Delta table in Unity Catalog Governed --> Indexed: Vector Search Indexed --> [*]
flowchart LR S0["Chunk"] --> S1["Delta table in Unity Catalog"] --> S2["Index builds from it"]
Key point: Store chunks in governed Delta tables.
Why the other options are wrong
- A. A temporary view that lasts for the notebook session
- A temporary view disappears when the session ends.
- B. A notebook's output cell
- An output cell is not storage for an index.
- C. A CSV file in a volume
- A CSV lacks the table features that a governed source provides.