Delta Lake is described as which of the following?

• A. A Spark API for exporting data into CSV files.
• B. A relational storage layer for Spark that supports Parquet-based tables.
• C. A data visualization library for Spark.
• D. A streaming ingestion tool.

Answer: (B)

Delta Lake acts as a storage layer for Spark that brings reliability to data lakes by layering ACID transactions on Parquet-based tables. It sits on top of your object store and uses a transaction log to track all changes, enabling atomic writes, consistent reads, and time travel (querying data as it existed at a past version). This design gives Spark SQL a familiar, relational-like experience when working with large Parquet datasets, including robust schema enforcement and the ability to perform upserts and merges.

It’s not a Spark API for exporting to CSV, not a data visualization library, and not a streaming ingestion tool.