To perform a one-time ingest of a small local CSV into a Fabric lakehouse with minimal administrative effort, which method is recommended?

To perform a one-time ingest of a small local CSV into a Fabric lakehouse with minimal administrative effort, which method is recommended?

• A. Dataflow Gen2 query
• B. Local file upload by using Lakehouse explorer
• C. Pipeline with the Copy data activity
• D. Spark notebook

Answer: (B)

For a one-time, small local CSV ingestion with minimal administrative effort, using the Lakehouse Explorer to upload the file directly is the best choice. This UI-based method lets you pick the local CSV and ingest it into a lakehouse table with little to no setup. It handles quick schema inference and creates the corresponding table without writing code or configuring pipelines, making it ideal for ad-hoc, simple loads. The other approaches introduce more overhead: a Dataflow Gen2 query is heavier, intended for transformations and repeatable processing; a pipeline with a copy data activity requires building and orchestrating a pipeline; a Spark notebook needs coding and managing a Spark compute environment. For a small, one-off load, the direct upload path keeps things streamlined and fast.

For a one-time, small local CSV ingestion with minimal administrative effort, using the Lakehouse Explorer to upload the file directly is the best choice. This UI-based method lets you pick the local CSV and ingest it into a lakehouse table with little to no setup. It handles quick schema inference and creates the corresponding table without writing code or configuring pipelines, making it ideal for ad-hoc, simple loads.

The other approaches introduce more overhead: a Dataflow Gen2 query is heavier, intended for transformations and repeatable processing; a pipeline with a copy data activity requires building and orchestrating a pipeline; a Spark notebook needs coding and managing a Spark compute environment. For a small, one-off load, the direct upload path keeps things streamlined and fast.