To interactively explore data in a file in the lakehouse using Spark, which option should you use?

• A. Switch to the SQL analytics endpoint mode
• B. Use a Spark shell
• C. Create a notebook
• D. Create a Dataflow (Gen2)

Answer: (C)

Notebooks provide an interactive workspace where you can run Spark code in small, runnable cells, inspect results as you go, and annotate your steps as you explore a file in the lakehouse. This setup supports an iterative workflow: load the data with Spark, apply transformations, preview outputs, refine queries, and visualize results all in one place. The other options serve different purposes—a Spark shell is a command-line REPL, which is great for quick ad-hoc experiments but isn’t as convenient for documenting steps or producing rich, repeatable explorations; switching to the SQL analytics endpoint mode emphasizes SQL queries through an endpoint rather than interactive Spark exploration; and a Dataflow (Gen2) is designed for building and orchestrating data pipelines, not for interactive data discovery.