When querying sales data files by using the SQL endpoint with data in S3, which file format should you use?

When querying sales data files by using the SQL endpoint with data in S3, which file format should you use?

• A. ORC
• B. Parquet
• C. CSV
• D. Delta format

Answer: (D)

Using Delta format is best because Delta Lake adds ACID transactions, a transaction log, and snapshot isolation on top of your S3 data. This gives a SQL endpoint a consistent view of your sales data even as new records arrive or existing ones are updated or deleted, which is common in analytics workloads. It also enables time travel to query historical states and supports schema evolution and upserts (MERGE), which are useful for sales corrections and incremental updates. Pure Parquet or ORC are fast columnar formats but lack the built-in transactional guarantees, while CSV is simple but inefficient and lacks reliability and data typing. Delta format combines the performance of a columnar format with the reliability and functionality needed for robust SQL querying on a data lake.

Using Delta format is best because Delta Lake adds ACID transactions, a transaction log, and snapshot isolation on top of your S3 data. This gives a SQL endpoint a consistent view of your sales data even as new records arrive or existing ones are updated or deleted, which is common in analytics workloads. It also enables time travel to query historical states and supports schema evolution and upserts (MERGE), which are useful for sales corrections and incremental updates. Pure Parquet or ORC are fast columnar formats but lack the built-in transactional guarantees, while CSV is simple but inefficient and lacks reliability and data typing. Delta format combines the performance of a columnar format with the reliability and functionality needed for robust SQL querying on a data lake.