Databricks Certified Data Engineer - Professional

Explanation

The correct answer is D. Funnel.

Understanding Data Ingestion Patterns:

• Sequence: This pattern involves processing data through a series of steps in a specific order, where each step depends on the output of the previous one. It's more about data transformation pipelines than combining multiple sources.

• Filter: This pattern involves removing unwanted data or selecting specific subsets of data based on certain criteria. It's about data reduction, not combining multiple sources.

• Fan-out: This pattern involves taking data from a single source and distributing it to multiple destinations or processing streams. It's the opposite of what's needed here.

• Funnel: This pattern is specifically designed to combine data from multiple sources into a single destination. In streaming scenarios, a funnel pattern collects data from various streaming sources and consolidates them into one refined table, which perfectly matches the requirement described in the question.

Why Funnel is Correct:

The question specifically asks about creating "a single refined table out of raw data from several streaming sources" - this is exactly what the funnel pattern accomplishes. It:

• Takes multiple input streams
• Processes and refines the data
• Outputs to a single destination table

This pattern is commonly used in streaming architectures where data from various sources (like Kafka topics, event streams, or different APIs) needs to be consolidated into a unified view for analytics or further processing.