Which of the following descriptions best fits Delta Lake?

• A. A Spark API for exporting data from a relational database into CSV files.
• B. A synchronization solution that replicates data between SQL Server and Spark.
• C. A machine learning library for Spark.
• D. A relational storage layer for Spark that supports tables based on Parquet files.

Answer: (D)

Delta Lake provides a storage layer that sits on top of data lakes to give you reliable, transactional access to data with a table-like model. It stores data as Parquet files and keeps a transactional log that enables ACID properties, schema management, and time travel, while allowing Spark to treat the data as relational tables. This makes the description of a relational storage layer for Spark that supports tables based on Parquet files the best fit. The other descriptions refer to exporting data to CSV, syncing data between SQL Server and Spark, or a machine learning library, which are not what Delta Lake is designed to provide.