Design and Implement Data Pipelines DP-750 Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   What is the primary benefit of using a Lakeflow Spark Declarative Pipeline for data transformation?

   • AIt provides a visual drag-and-drop interface for building pipelines
   • BIt eliminates the need for a compute cluster
   • CIt allows you to write arbitrary Python code for any transformation
   • DIt automatically manages dependencies and incremental processing based on the declared transformations
   Reveal Solution Discussion Select an answer first
2. 12expert · hard

   A team is evaluating whether to use a Lakeflow Spark Declarative Pipeline or a notebook for a new data pipeline. The pipeline must process streaming data, apply windowed aggregations, and write to a Delta table. The team also needs to implement custom stateful processing that is not easily expressed in SQL. They want to minimize infrastructure management and prefer a declarative approach where possible. What should they choose?

   • AUse a Lakeflow Spark Declarative Pipeline with a streaming source and a SQL window function.
   • BUse a Delta Live Tables pipeline with a streaming table and a custom Python function.
   • CUse a Lakeflow Spark Declarative Pipeline with a custom UDF for the stateful processing.
   • DUse a notebook with Structured Streaming, because it provides the flexibility to implement custom stateful processing and can still be scheduled as a Lakeflow job task.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   In a Lakeflow Spark Declarative Pipeline, how are data transformations typically defined?

   • AAs a set of SQL queries executed manually by an operator
   • BAs a series of functions that read from and write to Delta tables
   • CAs a collection of shell commands that invoke Spark submit
   • DAs a sequence of notebook cells that must be run in order
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A Lakeflow Job has three tasks: Task A reads raw data, Task B cleans the data, and Task C writes the cleaned data to a Delta table. The team wants to ensure that Task C only runs after Task B succeeds and that if Task B fails, the job stops without running Task C. What should they configure?

   • ASet Task B to depend on Task C, and set the job to continue on failure.
   • BSet Task C to depend on Task B, and set the job to stop on failure.
   • CRun all tasks in parallel and rely on the job scheduler to retry failed tasks.
   • DCombine all tasks into a single notebook and use conditional logic to skip Task C if Task B fails.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard · select all that apply

   A data pipeline processes a mix of valid and invalid records. The team wants to ensure that invalid records are quarantined and that the pipeline continues processing valid records. They also want to be able to monitor the number of invalid records. Which two strategies should they implement? (Select all that apply.)

   • AAdd a validation step that filters out invalid records and writes them to a separate Delta table.
   • BIgnore all errors and rely on the job scheduler to retry.
   • CUse the 'badRecordsPath' option when reading the data to route invalid records to a separate location.
   • DUse a try-except block to catch exceptions and log the invalid records.
   • ESet the job to stop on failure when any invalid record is encountered.
   Check answers Discussion Select an answer first