Question 36

You are designing ingestion into a Unity Catalog lakehouse for three sources. For each scenario, choose the **most appropriate ingestion approach** from the dropdown. The goal is to minimize long-term operational maintenance while keeping data fresh and governed by Unity Catalog. - **Scenario 1:** Ingest **Salesforce** opportunity and account objects with low-code setup, automatic incremental reads, and SCD type 2 history tracking, governed by Unity Catalog on serverless compute. - **Scenario 2:** Ingest **new JSON files arriving in an ADLS Gen2 external location**, where you need full control over complex transformations during ingestion using PySpark and Auto Loader. - **Scenario 3:** A team already has an existing enterprise **Azure Data Factory** orchestration and only needs to land copied files into storage, with Databricks reading them afterward. ```mermaid flowchart LR S1["Scenario 1:\nSalesforce SaaS,\nlow-code, SCD2"] --> D1{Dropdown 1} S2["Scenario 2:\nADLS JSON,\ncustom PySpark transforms"] --> D2{Dropdown 2} S3["Scenario 3:\nExisting ADF copy\norchestration"] --> D3{Dropdown 3} D1 -. options .-> O["Lakeflow Connect managed connector\n|\nAuto Loader in a notebook / pipeline\n|\nAzure Data Factory copy + Databricks read"] D2 -. options .-> O D3 -. options .-> O ```