BenchMark Quizzes by MDFT Pro

Identify Where Custom Steps Are Processed

You work for MDFT Pro, a well-known training agency. You have a Fabric tenant that contains two lakehouses, and you are building a dataflow that will combine data from the lakehouses.

The query settings from one of the queries in the dataflow are as follows:

Name: Students
Entity type: Custom
Applied steps:
  1. Source
  2. Navigation 1
  3. Capitalize each word
  4. Appended query
  5. Changed column type
  6. Added custom
  7. Filtered rows

Please complete the following statement:

The ‘Added custom’ step will be performed in ________

Choose the correct answer from the options below.

Explanations for each answer:

Each Lakehouse's Query Engine is incorrect. Lakehouse query engines only process steps that can be folded back to the source. Custom steps cannot be folded and therefore cannot run in the lakehouse query engines.
The Microsoft Power Query Engine is correct. Custom steps cannot be folded back to the data source, so they must be processed by the Power Query engine. The Power Query engine handles all transformation steps that cannot be pushed down to the source lakehouse, including custom formulas and functions.
The Source Lakehouse Query Engine is incorrect. The source lakehouse query engine only processes folded queries. Since custom steps cannot be folded, they must run in the Power Query engine instead.
The Target Lakehouse Query Engine is incorrect. The target lakehouse query engine is only used for writing the final output data. The 'Added custom' step is a transformation step that occurs during data reading, before any data reaches the target lakehouse.

Learn more about query folding in Fabric dataflows:

Query Folding in Fabric

Practice Exam

Identify Where Custom Steps Are Processed

Discuss this question on social media: