Skip to content
DP-700 Fabric Engineer Hub

Exam DP-700: Full Mock Exam 1 (Comprehensive Scenario Drill)

Exam DP-700: Full Mock Exam 1 (Comprehensive Scenario Drill)

[!NOTE] This is a model-authored full practice exam designed to test scenario readiness across all DP-700 objective domains.

Question 1

Domain: Ingest and Transform Data
You are building an event ingestion pipeline for clickstream analytics. You need to ingest web click events into OneLake, filter out bot traffic with sub-second latency, and land clean data in both a Delta table for historical BI and a real-time KQL database for active fraud monitoring. Which Fabric component should you implement?

  • A) Data Factory Copy Activity
  • B) Microsoft Fabric Eventstream with enhanced capabilities
  • C) Dataflows Gen2 with scheduled refresh
  • D) Fabric Notebook running on a scheduled Spark Starter Pool

Answer: B
Rationale: Fabric Eventstream with enhanced capabilities allows capturing high-throughput event streams, applying in-stream filtering operators (e.g. dropping bot user-agents), and routing to multiple destinations simultaneously (KQL Database and Lakehouse Delta table).

Question 2

Domain: Plan, Implement, and Manage Data Engineering Solutions
You manage a large enterprise data team. You want developers to be able to create, edit, and run PySpark notebooks, Data Factory pipelines, and Lakehouse items in a workspace, but they must NOT be allowed to add new users to the workspace or delete the workspace itself. Which workspace role should you assign to developers?

  • A) Admin
  • B) Member
  • C) Contributor
  • D) Viewer

Answer: C
Rationale: The Contributor role allows creating, editing, deleting, and executing items in a workspace, but does not permit managing workspace user permissions or deleting the workspace.

Question 3

Domain: Monitor and Optimize Data Engineering Solutions
A nightly PySpark batch ETL job that joins a 500GB fact table with a 20MB lookup table (dim_customer_segment) is taking 45 minutes to run. The Spark UI shows excessive shuffle read/write across all executors. How can you significantly optimize this join operation?

  • A) Partition the fact table by CustomerId.
  • B) Wrap the lookup table in broadcast(dim_customer_segment) within the join query.
  • C) Increase Spark node count from 2 to 64.
  • D) Convert the fact table to a CSV file.

Answer: B
Rationale: Broadcasting small dimension tables ($<100$ MB) distributes the lookup table to worker node memory, eliminating the expensive shuffle stage across the cluster for the 500GB fact dataset.

Question 4

Domain: Plan, Implement, and Manage Data Engineering Solutions
You need to provide access to 10 TB of historical telemetry logs stored in an external Amazon S3 bucket within your Fabric Lakehouse. The data must be queryable via Spark notebooks without replicating or copying the 10 TB of files into OneLake storage. What should you configure?

  • A) An Amazon S3 OneLake Shortcut in the Lakehouse Files/ directory.
  • B) A Data Factory Copy Activity running hourly.
  • C) Database Mirroring for Amazon S3.
  • D) An Azure ExpressRoute circuit.

Answer: A
Rationale: OneLake Shortcuts provide zero-copy virtualization, creating pointers directly to Amazon S3 buckets that allow Fabric compute engines to query external storage seamlessly.

Question 5

Domain: Ingest and Transform Data
You need to maintain slowly changing dimensions (SCD Type 1) in a Fabric Data Warehouse. New customer records and address updates arrive daily in a staging table. Which T-SQL statement should you use to perform atomic updates on existing records and inserts for new records?

  • A) INSERT INTO ... SELECT followed by DELETE
  • B) MERGE INTO TargetTable USING StagingTable ON TargetTable.CustomerId = StagingTable.CustomerId WHEN MATCHED THEN UPDATE ... WHEN NOT MATCHED THEN INSERT ...
  • C) ALTER TABLE TargetTable SWITCH PARTITION
  • D) SELECT * INTO TargetTable FROM StagingTable

Answer: B
Rationale: The MERGE statement in Fabric Data Warehouse provides atomic, single-statement upsert capabilities (updating matched records and inserting unmatched records).