DP-700 Practice Questions

Quick answer

These are 20 original DP-700 practice questions, each with the correct answer and a short explanation. They follow the three exam domains: implement and manage an analytics solution, ingest and transform data, and monitor and optimize an analytics solution.

• They are not real exam questions. Sharing real questions breaks Microsoft's exam agreement, and "dumps" are often wrong.
• Use them to find weak areas, then practise those areas hands-on in Microsoft Fabric.
• Finish with Microsoft's free official practice assessment before you book.

How to use these questions

1. Answer all 20 without looking at the explanations.
2. Score yourself by domain: 7 questions for Domain 1, 7 for Domain 2 and 6 for Domain 3.
3. For every wrong answer, build the scenario yourself in a Fabric workspace.
4. Review the full topic list in our DP-700 syllabus and study guide.

1. Domain 1: Implement and manage an analytics solution

Short answer: These questions test workspace settings, Git and deployment pipelines, security and orchestration choices.

Q1. Your team wants every change to Fabric notebooks and pipelines tracked in a repository, with branches and pull requests. What should you configure?

• A. A deployment pipeline
• B. Git integration for the workspace
• C. Sensitivity labels
• D. A OneLake shortcut

Q2. You must move a Lakehouse, notebooks and a semantic model from development to test and then production with as little manual work as possible. What should you use?

• A. Deployment pipelines
• B. Dataflow Gen2
• C. Mirroring
• D. Manual export and import

Q3. Sales analysts query a Fabric Warehouse. Each analyst must see only the rows for their own region. What should you implement?

• A. Dynamic data masking
• B. Column-level security
• C. Row-level security
• D. Workspace Viewer role

Q4. Support staff must be able to query a customer table, but they should see only a partly hidden email address. What is the simplest option?

• A. Row-level security
• B. Dynamic data masking on the email column
• C. Remove the email column
• D. Sensitivity labels

Q5. A finance user needs read access to one Warehouse but must not see anything else in the workspace. What should you do?

• A. Add the user as a workspace Viewer
• B. Add the user as a workspace Contributor
• C. Share only the Warehouse item with read permission
• D. Give the user Fabric administrator rights

Q6. A business analyst who knows Power Query needs to clean and reshape a CSV file every day, without writing code. Which tool fits best?

• A. Notebook
• B. Dataflow Gen2
• C. T-SQL stored procedure
• D. KQL queryset

Q7. A pipeline must pass today's date into a notebook so the notebook loads only that day's data. How should you set this up?

• A. Hard-code the date in the notebook
• B. Mark a notebook cell as the parameter cell and pass the value from the pipeline's Notebook activity
• C. Use a OneLake shortcut
• D. Use a sensitivity label

2. Domain 2: Ingest and transform data

Short answer: These questions test loading patterns, shortcuts and mirroring, choosing the right store and engine, and transformations in PySpark, T-SQL and KQL.

Q8. Your files already live in Azure Data Lake Storage Gen2. You want to query them from a Fabric Lakehouse without copying the data. What should you create?

• A. A OneLake shortcut
• B. A copy activity in a pipeline
• C. A Dataflow Gen2
• D. A mirrored database

Q9. You need a continuously updated copy of an operational Azure SQL Database in Fabric with minimal setup and no custom pipelines. What should you use?

• A. A scheduled full-load pipeline
• B. Mirroring
• C. A OneLake shortcut
• D. An Eventstream

Q10. A table has 500 million rows and grows by about 2 million rows a day. Loading everything every night takes too long. What loading pattern should you use?

• A. Full load every night
• B. Incremental load using a watermark column
• C. Manual uploads
• D. Delete and recreate the table daily

Q11. Customer records sometimes arrive late with updated addresses. You must update existing rows and insert new ones in a Warehouse dimension table in one statement. Which T-SQL statement should you use?

• A. TRUNCATE TABLE
• B. INSERT INTO only
• C. MERGE
• D. SELECT INTO

Q12. IoT sensors send thousands of events per second. Analysts need to query the latest readings within seconds. Which combination fits best?

• A. Pipeline copy every hour into a Warehouse
• B. Eventstream into an Eventhouse, queried with KQL
• C. Dataflow Gen2 into a Lakehouse once a day
• D. CSV export to OneLake

Q13. In an Eventhouse, you need the number of events per device in fixed 5-minute buckets. Which KQL pattern is correct?

• A. Events | summarize count() by DeviceId, bin(Timestamp, 5m)
• B. Events | where count() > 5m
• C. Events | project bin(Timestamp)
• D. SELECT COUNT(*) FROM Events GROUP BY 5

Q14. A data engineering team prefers T-SQL and needs multi-table inserts, updates and deletes with transactions. Which Fabric item should store the data?

• A. Lakehouse, through its SQL analytics endpoint
• B. Warehouse
• C. Eventhouse
• D. A OneLake shortcut

3. Domain 3: Monitor and optimize an analytics solution

Short answer: These questions test monitoring, error handling and performance tuning across pipelines, notebooks, Lakehouse tables and shortcuts.

Q15. You need one place to see recent runs and failures of pipelines, notebooks and dataflows across the workspaces you can access. Where should you look?

• A. The Monitor hub
• B. OneLake file explorer
• C. Sensitivity labels
• D. The deployment pipeline

Q16. Queries on a Lakehouse table have slowed down. The table now contains thousands of very small Parquet files. What should you run first?

• A. VACUUM
• B. OPTIMIZE to compact the small files
• C. DROP TABLE
• D. Add a sensitivity label

Q17. Old, unreferenced files from earlier versions of a Delta table are using a lot of storage. Which command removes them?

• A. OPTIMIZE
• B. VACUUM
• C. MERGE
• D. ANALYZE

Q18. The operations team wants an email whenever a nightly pipeline fails. What is the simplest approach inside the pipeline?

• A. Add an Office 365 Outlook (or Teams) activity on the failure path of the last activity
• B. Ask someone to check the pipeline every morning
• C. Add a OneLake shortcut
• D. Convert the pipeline to a Dataflow Gen2

Q19. A notebook that reads through a OneLake shortcut suddenly fails with an access error, although the notebook code has not changed. What should you check first?

• A. The Spark pool size
• B. Permissions on the shortcut's target location
• C. The sensitivity label on the notebook
• D. The deployment pipeline stage

Q20. After deploying a notebook from development to test, it fails with "table not found". It works in development. What is the most likely cause?

• A. The notebook is still attached to the development Lakehouse, or no default Lakehouse is set in test
• B. KQL is not enabled
• C. The exam sandbox is closed
• D. The table has too many columns

4. How did you score?

Short answer: 17 or more correct means you are close to ready. Below 14, go back to hands-on practice in your weakest domain before booking the exam.

5. Free official DP-700 practice test

Short answer: Microsoft offers a free DP-700 practice assessment on Microsoft Learn with questions in the style, wording and difficulty of the real exam, plus an exam sandbox to try the interface.

• Practice assessment: "Practice for the exam" on the Fabric Data Engineer Associate page
• Exam sandbox: try the exam interface
• Booking, fee (₹4,865 in India), retakes and renewal: our Azure data engineer certification guide

Why you should avoid DP-700 dumps

1. Sharing real exam questions breaks the Microsoft exam agreement, and certifications can be revoked.
2. Dump answers are often wrong or outdated, especially after the October 2026 update.
3. Interviewers test the judgement behind the certificate, which dumps don't teach.

6. How Brolly Academy helps

• Azure Data Engineer Training in Hyderabad: taught by Dinesh (9+ years), with Azure Data Factory, Synapse, Databricks, PySpark, SQL and Python, real-time projects, mock interviews and placement support, from ₹9,999.
• Microsoft Fabric Course in Hyderabad: a one-month Fabric program from ₹9,999.
• Classroom, online and hybrid batches in English and Telugu, with a free demo. Explore all courses at Brolly Academy.

Frequently asked questions

Sources and further reading

• Microsoft Learn — Study guide for Exam DP-700
• Microsoft Learn — Microsoft Certified: Fabric Data Engineer Associate
• Microsoft Learn — Exam duration and exam experience
• Microsoft Learn — OneLake shortcuts
• Microsoft Learn — Mirroring in Microsoft Fabric
• Apache Software Foundation — Apache Spark