CertifyCloudxAce every cloud certification.

DP-600 sample questions with answers

10 free practice questions for the Microsoft Certified: Fabric Analytics Engineer Associate exam. Try each one, then open the answer to see why the right option wins and every other option loses.

Question 1Maintain a data analytics solution

Adventure Works maintains Dev, Test, and Prod workspaces on the same Fabric capacity. Every change must be reviewed in a pull request before it can reach Test, and promotion between environments must not require re-authoring content. The team wants the smallest number of moving parts. Which approach should you recommend?

  1. A.

    Connect all three workspaces to three long-lived branches and promote content by merging branches, without a deployment pipeline.

  2. B.

    Connect the Prod workspace to the main branch and have developers commit from Prod after each deployment so that the repository always matches production.

  3. C.

    Connect only the Dev workspace to the repository, require pull requests into the main branch, and use a deployment pipeline to promote Dev to Test to Prod.

  4. D.

    Skip Git integration and export each item from Dev, then import it into Test and Prod after the review meeting.

Show answer

Answer: C

Git on the development workspace gives code review through pull requests, and a deployment pipeline promotes the reviewed content forward without rebuilding it in each environment.

  • A. Three connected workspaces and branch merges add sync overhead and give up deployment rules and the pipeline comparison view.
  • B. Committing from Prod makes production the source of truth, so changes made there are never reviewed before they take effect.
  • C. Git on Dev supplies the pull request gate while the pipeline promotes the same reviewed artifacts forward with rules and auto-binding.
  • D. Manual export and import has no version history, cannot enforce a pull request, and loses automatic rebinding of dependent items.
Question 2Maintain a data analytics solution

Litware must let five regional teams build their own reports on a single governed sales model without each team owning a copy of it. Litware must also hand an external partner a starting file that carries the corporate theme, the standard page layout, and the query definitions, but no data at all. Which two actions should you perform? Each correct answer presents a complete solution. (Choose TWO.)

Choose 2.

  1. A.

    Give each regional team a copy of the .pbix file so that every team can refresh its own model.

  2. B.

    Publish the semantic model to a workspace, grant each team the Build permission on it, and have them create reports with a live connection from the OneLake catalog.

  3. C.

    Send the partner a Power BI data source (.pbids) file that contains the theme and the page layout.

  4. D.

    Export the semantic model from the service as a .pbix file and email it to the partner with the data removed.

  5. E.

    Save the report as a Power BI template (.pbit) file and send it to the partner.

Show answer

Answer: B, E

Build permission plus a live connection lets many teams report on one shared model, and a .pbit template carries layout, theme, and queries without any data.

  • A. Copying the .pbix gives every team a separate model and refresh, which defeats the single governed model requirement.
  • B. Build permission on a published model is what lets other teams author reports against it with a live connection instead of copying it.
  • C. A .pbids file contains only connection information and launches Get data; it can carry neither a theme nor a page layout.
  • D. A .pbix downloaded from the service still contains the imported data, and XMLA-modified models cannot be downloaded at all.
  • E. A .pbit carries visuals, theme, model schema, and queries but no data, which matches the partner requirement exactly.
Question 3Maintain a data analytics solution

A workspace Admin at Litware opens Workspace settings > Git integration to connect the Marketing workspace to a GitHub repository, but only Azure DevOps is offered as a provider. Other workspaces in the tenant are already connected to Azure DevOps. What should a Fabric administrator do?

  1. A.

    Enable the setting that allows exporting items to Git repositories in other geographical locations.

  2. B.

    Assign the Marketing workspace to an F64 capacity, because GitHub requires a larger capacity than Azure DevOps.

  3. C.

    Give the workspace Admin the Contributor role in the GitHub organization that owns the repository.

  4. D.

    Enable the tenant setting Users can sync workspace items with GitHub repositories, which is disabled by default.

Show answer

Answer: D

The GitHub integration tenant switch is disabled by default and must be enabled before workspaces can connect to GitHub.

  • A. The cross-geo setting applies to Azure DevOps repos in other regions and isn't enforced for GitHub at all.
  • B. Git integration needs a Fabric capacity, but no provider requires a particular SKU size.
  • C. Git-side permissions matter after the provider is available; they don't make the provider appear in Fabric.
  • D. Correct. The GitHub switch is disabled by default and must be enabled, for the organization or specific groups, before GitHub appears.
Question 4Maintain a data analytics solution

Contoso has enabled sensitivity labels in Fabric. A data engineer who holds the Contributor role and a Power BI Pro license opens a lakehouse's sensitivity flyout, but the Confidential label is greyed out. Other engineers can apply it. What is the most likely cause?

  1. A.

    The lakehouse must be endorsed as Certified before a Confidential label can be applied to it.

  2. B.

    Only workspace Admins can apply sensitivity labels to lakehouses and other non-Power BI items.

  3. C.

    The workspace must run on an F64 or larger capacity for any sensitivity labels to be applied.

  4. D.

    The Confidential label isn't published to the engineer in a Microsoft Purview label policy.

Show answer

Answer: D

A user can apply a label only if they have edit permission, a Pro or PPU license, and the label is published to them in Purview.

  • A. Endorsement and labeling are independent features; neither is a prerequisite for the other.
  • B. Applying a label requires edit permission on the item, which a Contributor has; Admin isn't required.
  • C. Capacity size doesn't gate labeling, and other engineers in the same workspace can apply the label.
  • D. Correct. Being allowed to apply labels in Fabric isn't enough; the label must also be published to the user in Purview.
Question 5Maintain a data analytics solution

A developer at Relecloud saves a report as a Power BI project (.pbip) inside a local Git repository. The security team asks which project file holds a cached copy of the model's data so that it never reaches the remote repository. Which file should you name?

  1. A.

    definition.pbir in the report folder

  2. B.

    definition.pbism in the semantic model folder

  3. C.

    The .platform file in each item folder

  4. D.

    .pbi\cache.abf in the semantic model folder

Show answer

Answer: D

cache.abf holds the local model and data cache and is excluded by the .gitignore that Power BI Desktop creates.

  • A. definition.pbir holds the report definition and its reference to the semantic model, not data.
  • B. definition.pbism holds the overall semantic model definition and format version, not data.
  • C. The .platform file holds properties that link the item to Git, not a copy of data.
  • D. Correct. cache.abf is an Analysis Services backup containing a local cached copy of the model and data, and it's git-ignored by default.
Question 6Maintain a data analytics solution

Tailwind Traders publishes a shared semantic model on an F64 capacity. Twenty store analysts who hold only Microsoft Fabric free licenses have Build permission and want to create and publish their own reports on it in a team workspace. What do they need?

  1. A.

    The Viewer role in the model's workspace, which grants authoring rights to free users on F64.

  2. B.

    Nothing more, because Build permission on an F64 capacity lets free users author and publish anywhere.

  3. C.

    Power BI Pro or Premium Per User licenses, because free users can author reports on semantic models only in My workspace.

  4. D.

    A Fabric trial capacity for each analyst, because trials include authoring rights for semantic models.

Show answer

Answer: C

Free users can author on shared semantic models only in My workspace; publishing to team workspaces needs Pro or PPU.

  • A. Viewer governs viewing; authoring and publishing Power BI content in shared workspaces still requires Pro or PPU.
  • B. F64 lets free users view content with the Viewer role; it doesn't license them to author Power BI items outside My workspace.
  • C. Correct. Users without Pro or PPU can use semantic models for report authoring only when the models reside in their My workspace.
  • D. A trial capacity doesn't change the per-user license needed to author Power BI items in shared workspaces.
Question 7Maintain a data analytics solution

Woodgrove Bank wants Power BI content in its deployment pipeline to move from Test to Production only after an approval gate in an Azure DevOps release, with no manual clicks in the Fabric portal. What should the release run?

  1. A.

    A task that exports each Test item as a .pbix file and republishes it to the Production workspace

  2. B.

    A task that runs a TMSL createOrReplace script against each Production model through the XMLA endpoint

  3. C.

    A task that commits the Test workspace to Git so that Production updates itself from the repository

  4. D.

    A task that signs in with a service principal and calls the deployment pipelines REST API to deploy from Test

Show answer

Answer: D

Automate pipeline promotions by calling the deployment pipelines REST APIs from Azure DevOps with a service principal or user.

  • A. Republishing bypasses pairing, rules, and deployment history and isn't the pipeline's deployment mechanism.
  • B. XMLA scripting would bypass the pipeline and wouldn't deploy reports; it isn't pipeline deployment.
  • C. Committing Test doesn't deploy anything to Production, which isn't synced from that commit.
  • D. Correct. The deployment pipelines APIs support deploy all and selective deploy, and Azure DevOps can call them with a service principal.
Question 8Maintain a data analytics solution

An analyst at Woodgrove Bank was granted Read all with SQL analytics endpoint on a lakehouse named Loans-LH. She can query Loans-LH from SQL Server Management Studio with the SQL connection string, but when she opens Loans-LH by its lakehouse URL in the Fabric portal, she can't browse the tables. What should you do?

  1. A.

    Grant her Build permission on a semantic model that is created from the Loans-LH tables.

  2. B.

    Remove Read all with SQL analytics endpoint and share Loans-LH again with no additional permissions.

  3. C.

    Also grant her the Read all with Apache Spark permission on Loans-LH.

  4. D.

    Grant her the Execute Apache Spark permission so that the Livy endpoint can list the tables for her.

Show answer

Answer: C

Users with only the SQL analytics endpoint permission can't read data through the lakehouse URL; add Read all with Apache Spark.

  • A. Build governs report authoring on a semantic model and doesn't change what she can browse in the lakehouse.
  • B. Read alone lets her open the item but not see data, so this would remove access instead of adding it.
  • C. Correct. Reading lakehouse data through the lakehouse URL and explorer requires ReadAll in addition to the SQL permission.
  • D. Execute Apache Spark allows running Spark jobs through the Livy endpoint; it is not the read permission the explorer uses.
Question 9Maintain a data analytics solution

At Tailwind Traders, an analyst holds the Viewer role in the Ops workspace. She can query the SalesLH lakehouse through its SQL analytics endpoint, but a Spark notebook in her own workspace that reads the same Delta tables through OneLake fails with an access error. What is the cause?

  1. A.

    The Viewer role grants T-SQL read access through the SQL analytics endpoint but not the ReadAll permission that Spark and the OneLake APIs require.

  2. B.

    Spark notebooks can read only lakehouses that are stored in the same workspace as the notebook, so the notebook must be moved into Ops.

  3. C.

    Her Microsoft Fabric free license allows T-SQL queries but blocks every Spark session until a Power BI Pro license is assigned to her.

  4. D.

    The SQL analytics endpoint keeps a private copy of the tables, so the Delta files in OneLake haven't been written yet for Spark to read.

Show answer

Answer: A

Viewer includes T-SQL read access to lakehouse data but not ReadAll, so Spark and OneLake API reads fail while SQL endpoint queries succeed.

  • A. Correct. Viewers can read lakehouse data with T-SQL (ReadData) but the ReadAll path used by Spark, OneLake APIs, and the lakehouse explorer is reserved for Contributor and above.
  • B. Notebooks can read lakehouses in other workspaces through OneLake paths or shortcuts; location is not what blocks her.
  • C. Spark access to non-Power BI Fabric items is not gated by a Pro license; the failure comes from the permission level of her role.
  • D. The SQL analytics endpoint reads the same Delta files in OneLake; there is no separate copy that Spark would be waiting for.
Question 10Maintain a data analytics solution

Northwind Traders connects an existing workspace to a GitHub repository. Several semantic models in the workspace carry sensitivity labels, and committing them to Git is blocked. The team accepts that Git won't store the labels. What should a Fabric administrator do so the team can commit these items?

  1. A.

    Enable the tenant setting that lets users export items with applied sensitivity labels to Git repositories.

  2. B.

    Save each semantic model as a Power BI project so that its sensitivity label is stored in the text definition files that Git tracks.

  3. C.

    Change the Git provider to Azure DevOps, which stores sensitivity labels as repository metadata.

  4. D.

    Remove the labels, commit, and then reapply the labels after every commit to keep items protected.

Show answer

Answer: A

Git can't carry sensitivity labels; an admin setting allows items with applied labels to be exported to Git without the label.

  • A. Correct. Git doesn't support labels, and exporting labeled items can be disabled; this admin setting allows committing them without the label.
  • B. Sensitivity labels aren't supported with Power BI projects, so PBIP files can't carry them either.
  • C. Neither Git provider stores Fabric sensitivity labels; the restriction comes from Fabric, not the provider.
  • D. Stripping and reapplying labels on every commit is unsafe and unnecessary when an admin setting allows the export.

Keep going with 502 more DP-600 questions

Free papers every day, in the real exam formats, with progress by exam domain. Unlock every paper and timed mock exam when you are ready.

Practise freeSee all DP-600 papers
DP-600 sample questions with answers (10 free) · CertifyCloudx