Microsoft Fabric Analytics Engineer DP-600 Practice Test

DP-600 practice test: what's covered

• Lakehouse
• Warehouse
• Dataflows
• T-SQL
• Spark
• Power BI
• DAX

Free DP-600 sample questions

1. An analytics engineer must group several related workspaces for the Finance organization so that data owners can manage discoverability and governance for their subject area without admin control over the whole tenant. Which Microsoft Fabric construct should be used?

   • A. A Fabric domain that bundles the workspaces under a Finance namespace
   • B. A single workspace with separate folders for each team
   • C. A dedicated capacity assigned only to the Finance tenant
   • D. A deployment pipeline linking the Finance workspaces
   Show answer

   Correct: A. Domains are a tenant-level way to logically group multiple workspaces by business area and delegate governance to domain admins, which is exactly the discoverability and ownership need described. A single workspace with folders cannot span teams or delegate governance. Capacity governs compute, not logical grouping. Deployment pipelines only promote content between dev/test/prod stages.

2. Your team stores all Fabric items in OneLake and wants reusable, governed access to a folder of Parquet files that physically live in an external Azure Data Lake Storage Gen2 account, without copying the data into the lakehouse. Which feature meets this requirement?

   • A. A OneLake shortcut pointing to the ADLS Gen2 location
   • B. A Dataflow Gen2 that ingests the Parquet files nightly
   • C. A Data pipeline Copy activity into a managed Delta table
   • D. A mirrored database connection to the storage account
   Show answer

   Correct: A. OneLake shortcuts create a virtual reference to data in another location (ADLS Gen2, S3, or another lakehouse) so it can be queried in place without duplication. A Dataflow or pipeline Copy activity would physically move and store a second copy. Mirroring targets operational databases, not a folder of Parquet files in storage.

3. An analytics engineer is preparing a Fabric workspace for production and must enforce least-privilege access while still letting report consumers view dashboards. Which workspace role assignments correctly satisfy these goals? (Choose TWO)

   • A. Grant report consumers the Viewer role so they can read items but not edit them
   • B. Grant data engineers the Contributor role so they can create and edit items
   • C. Grant report consumers the Admin role so dashboards always render
   • D. Grant data engineers the Viewer role so they can build pipelines
   • E. Grant every user the Member role to simplify permission management
   Show answer

   Correct: A and B. Viewer is read-only and is the correct least-privilege role for consumers, while Contributor lets engineers create and edit items without granting workspace management rights. Giving consumers Admin massively over-permissions them. Viewer cannot build pipelines, so it is wrong for engineers. Blanket Member access violates least privilege by giving everyone broad edit rights.

4. A data team uses a Fabric-enabled capacity and notices that interactive Power BI queries slow down dramatically every afternoon when large Spark notebook jobs run on the same capacity. They want background jobs to keep running without throttling interactive users. Which capacity feature explains this behavior and the built-in mitigation?

   • A. Smoothing and bursting absorb short spikes and spread compute over time to limit throttling
   • B. Row-level security automatically reduces the rows Spark must scan during peaks
   • C. DirectLake mode disables Spark whenever interactive queries are detected
   • D. Per-user licensing isolates each user onto a separate dedicated capacity
   Show answer

   Correct: A. Fabric capacities use bursting to temporarily exceed the purchased compute and smoothing to average usage across time, which softens contention between background and interactive workloads. Row-level security filters report rows and has nothing to do with Spark scan cost. DirectLake never disables Spark. Per-user licensing does not create isolated capacities; all workloads share the assigned capacity.

5. An organization must apply consistent classification and protection to sensitive data across Fabric lakehouses, semantic models, and exported Power BI files. Which Microsoft capability should the analytics engineer integrate to label and protect this content?

   • A. Microsoft Purview sensitivity labels applied to Fabric items
   • B. Workspace Git integration with branch protection rules
   • C. A dedicated F64 capacity reserved for sensitive workloads
   • D. DirectQuery storage mode enforced on every semantic model
   Show answer

   Correct: A. Purview sensitivity labels classify and protect Fabric content and flow through to exports such as Excel and PDF, providing consistent, persistent protection. Git integration manages source control, not data classification. A larger capacity changes compute, not labeling. DirectQuery is a storage mode and does not classify or encrypt content.

Last updated: 2026-10-19

DP-600 practice test FAQ

Related practice tests

• Microsoft DP-700 practice test
• Microsoft PL-300 practice test
• Microsoft DP-900 practice test
• Microsoft DP-300 practice test
• Databricks Data Engineer Associate practice test
• SnowPro Core COF-C02 practice test