Answer: Lakehouse supports schema enforcement and evolution, traditional data warehouses lack schema evolution.

✅ E. Lakehouse supports schema enforcement and evolution, traditional data warehouses lack schema evolution. This is the most significant advantage in this scenario. Traditional data warehouses typically enforce a rigid schema-on-write approach. This means the structure of the data must be defined before it is loaded. Handling highly unstructured and frequently changing data in such a system becomes cumbersome, requiring extensive ETL processes and schema modifications that can be disruptive. Lakehouses, on the other hand, embrace a schema-on-read approach for the data lake portion and offer capabilities for schema evolution as data is processed and transformed within the managed layer (often using Delta Lake on Databricks). This allows you to land the raw, unstructured data easily and then define and evolve schemas as your understanding of the data and your analytical needs change. ❌ A. Lakehouse supports SQL While Lakehouses do support SQL for querying and analysis (often through engines like Spark SQL), modern data warehouses also extensively support SQL. This is not a differentiating factor that makes a Lakehouse better for handling unstructured and evolving data. ❌ B. Lakehouse supports ACID ACID (Atomicity, Consistency, Isolation, Durability) properties are crucial for data reliability and transactional integrity. Modern Lakehouse architectures, especially when built on technologies like Delta Lake, do support ACID properties. However, many modern data warehouses also provide strong ACID compliance. This is an important feature of a Lakehouse but not the primary reason it’s superior for unstructured and evolving data. ❌ C. Lakehouse enforces data integrity Both Lakehouses and data warehouses aim to ensure data integrity. Data warehouses typically enforce integrity through schema constraints, data types, and ETL processes. Lakehouses achieve data integrity through features offered by their underlying technologies (like Delta Lake’s support for schema enforcement, data validation, and transaction management). While Lakehouses provide mechanisms for data integrity, it’s not the core reason they are better suited for unstructured and evolving data compared to the schema flexibility aspect. ❌ D. Lakehouse supports primary and foreign keys like a data warehouse While Lakehouse architectures are evolving, and technologies like Delta Lake are adding support for constraints, traditional relational data warehouses have historically had more mature and robust support for primary and foreign key constraints for enforcing relationships and data integrity at the database level. This is an area where data warehouses have traditionally been stronger. The primary advantage of a Lakehouse for unstructured and evolving data lies in its flexibility with schema.