DP-900 sample exam questions

15 exam-style questions for Microsoft DP-900: Azure Data Fundamentals, with answers and explanations. Take them as a timed practice exam first if you want a score out of 1000.

Take the free practice exam DP-900 exam guide
  1. Question 1Single choiceNon-relational data on Azure
    Which type of database contains nodes and edges?
    1. A
      graph
    2. B
      key/value
    3. C
      columnar
    4. D
      time series
    Show answer

    Correct answer: A

    Why

    A graph database stores entities as nodes and the relationships between them as edges, which makes traversing connections efficient. Key/value, columnar and time series databases do not model data as nodes and edges.

  2. Question 2Single choiceOther
    When can you use an Azure Resource Manager template?
    1. A
      to automate the creation of an interdependent group of Azure resources in a repeatable way
    2. B
      to apply Azure policies for multi-tenant deployments
    3. C
      to provision Azure subscriptions
    4. D
      to control which services and feature administrators and developers can deploy from the Azure portal
    Show answer

    Correct answer: A

    Why

    ARM templates are declarative JSON files that deploy an interdependent group of resources consistently and repeatably as infrastructure as code. Azure Policy governs and restricts what can be deployed, and templates do not provision subscriptions.

  3. Question 3Single choiceRelational data on Azure
    Which clause should you use in a SELECT statement to combine rows in one table with rows in another table?
    1. A
      SET
    2. B
      VALUES
    3. C
      KEY
    4. D
      JOIN
    Show answer

    Correct answer: D

    Why

    JOIN combines rows from two or more tables based on a related column, such as a customer ID. SET is used in UPDATE, VALUES supplies rows for INSERT, and KEY is not a SELECT clause.

  4. Question 4Single choiceNon-relational data on Azure
    Your company is designing an application that will write a high volume of JSON data and will have an application-defined schema.
    Which type of data store should you use?
    1. A
      columnar
    2. B
      key/value
    3. C
      document
    4. D
      graph
    Show answer

    Correct answer: C

    Why

    A document database stores JSON documents whose structure is defined by the application rather than by a fixed database schema, and handles high write volumes, as with Cosmos DB for NoSQL. Key/value stores treat values as opaque, while columnar and graph stores target analytics and relationships.

  5. Question 5Multiple responseAnalytics workloads on Azure
    Which two activities can be performed entirely by using the Microsoft Power BI service without relying on Power BI Desktop? Each correct answer presents a
    complete solution.
    NOTE: Each correct selection is worth one point.
    1. A
      report and dashboard creation
    2. B
      report sharing and distribution
    3. C
      data modeling
    4. D
      data acquisition and preparation
    Show answer

    Correct answer: A, B

    Why

    In the Power BI service you can create reports from published semantic models, pin visuals to build dashboards, and share and distribute content through workspaces and apps. Rich data modeling and full Power Query data preparation are mainly done in Power BI Desktop.

  6. Question 6Single choiceAnalytics workloads on Azure
    What can be used with native notebook support to query and visualize data by using a web-based interface?
    1. A
      Azure Databricks
    2. B
      pgAdmin
    3. C
      Microsoft Power BI
    Show answer

    Correct answer: A

    Why

    Azure Databricks provides a web-based workspace with native notebooks for querying data with Spark SQL, Python or Scala and visualizing results. pgAdmin is a PostgreSQL admin tool, and Power BI is a reporting tool rather than a notebook environment.

  7. Question 7Single choiceOther
    Which database transaction property ensures that individual transactions are executed only once and either succeed in their entirety or roll back?
    1. A
      atomicity
    2. B
      durability
    3. C
      isolation
    4. D
      consistency
    Show answer

    Correct answer: A

    Why

    Atomicity means a transaction is a single indivisible unit: it either completes in full or is rolled back completely. Durability preserves committed changes, isolation separates concurrent transactions, and consistency keeps data valid.

  8. Question 8Single choiceCore data concepts
    A bank has a system that manages financial transactions.
    When transferring money between accounts. the system must never retrieve a value for the source account that reflects the balance before the transfer and a
    value for the destination account that reflects the balance after the transfer.
    Of which ACID semantic is this an example?
    1. A
      atomicity
    2. B
      durability
    3. C
      consistency
    4. D
      isolation
    Show answer

    Correct answer: D

    Why

    Isolation ensures concurrent transactions do not see each other's intermediate states, so no reader sees the source debited before the transfer completes and the destination credited after it. Atomicity is all-or-nothing, consistency keeps valid states, and durability preserves committed data.

  9. Question 9Single choiceCore data concepts
    A data engineer is responsible for which task?
    1. A
      Explore data to identify trends.
    2. B
      Implement policies, tools, and processes for backup and recovery plans.
    3. C
      Design and build analytical models.
    4. D
      Design and implement data stores for analytical workloads.
    Show answer

    Correct answer: D

    Why

    Data engineers design and build data stores and pipelines for analytical workloads, such as data lakes and warehouses. Exploring trends and building analytical models are data analyst tasks, and backup and recovery policies are database administrator tasks.

  10. Question 10HotspotNon-relational data on Azure
    HOTSPOT

    Select the answer that correctly completes the sentence.

    In a graph database, edges specify relationships between ___column families / facts / vertexes / tables
    Show answer
    • In a graph database, edges specify relationships between ___ vertexes
    Why

    A graph database stores entities as vertices (nodes) and the relationships between them as edges, as in Azure Cosmos DB for Apache Gremlin. Column families belong to column-family stores, and facts and tables to relational and warehouse models.

  11. Question 11HotspotRelational data on Azure
    HOTSPOT

    For each of the following statements, select Yes if the statement is true. Otherwise, select No.

    NOTE: Each correct selection is worth one point.

    The Cool access tier is the default tier of an Azure Storage account.Yes / No
    Append blobs are optimized for random read and write operations.Yes / No
    Retrieving data from the Archive access tier takes hours to complete.Yes / No
    Show answer
    • The Cool access tier is the default tier of an Azure Storage account. No
    • Append blobs are optimized for random read and write operations. No
    • Retrieving data from the Archive access tier takes hours to complete. Yes
    Why

    Hot is the default access tier for a new storage account (Cool can be chosen instead). Append blobs are optimized for append operations such as logging, while page blobs support random read and write. Archive blobs must be rehydrated, which can take up to 15 hours.

  12. Question 12HotspotCore data concepts
    HOTSPOT

    Select the answer that correctly completes the sentence.

    Online analytical processing (OAP) systems are particularly suitable for analytical workloads because data is ___limited in size. / optimized for transactions. / pre-aggregated. / read-only.
    Show answer
    • Online analytical processing (OAP) systems are particularly suitable for analytical workloads because data is ___ pre-aggregated.
    Why

    OLAP models pre-aggregate measures across dimension hierarchies, so analytical queries return quickly without recalculating from detail rows. OLAP data is not limited in size or optimized for transactions, and being read-only is not what makes it fast for analysis.

  13. Question 13Drag and dropAnalytics workloads on Azure
    DRAG DROP

    Match the data processing languages to the appropriate real-time analytics services.

    To answer, drag the appropriate language from the column on the left to its service on the right. Each language may be used once, more than once, or not at all.

    NOTE: Each correct match is worth one point.

    Move each item to the matching row:

    • JavaScript
    • KQL
    • Python
    • SQL
    • Azure Data Explorer
    • Azure Stream Analytics
    • Azure Synapse Analytics
    Show answer
    • Azure Data Explorer KQL
    • Azure Stream Analytics SQL
    • Azure Synapse Analytics SQL
    Why

    Azure Data Explorer queries data with Kusto Query Language (KQL). Azure Stream Analytics jobs use a SQL-based query language, and Azure Synapse Analytics uses T-SQL in its SQL pools (it also supports Python in Spark, but SQL is the intended match). JavaScript is only used for Stream Analytics user-defined functions.

  14. Question 14Drag and dropCore data concepts
    DRAG DROP

    Match the job roles to the appropriate tasks.

    To answer, drag the appropriate role from the column on the left to its task on the right. Each role may be used once, more than once, or not at all.

    NOTE: Each correct match is worth one point.

    Move each item to the matching row:

    • Data analyst
    • Data engineer
    • Database administrator
    • Database backup and recovery
    • Enabling advanced analytic capabilities through reports
    • Monitoring data pipelines
    Show answer
    • Database backup and recovery Database administrator
    • Enabling advanced analytic capabilities through reports Data analyst
    • Monitoring data pipelines Data engineer
    Why

    Database administrators handle backup and recovery, data analysts deliver insights and analytic capabilities through reports and visualizations, and data engineers build, run and monitor data pipelines.

  15. Question 15Single choiceRelational data on Azure
    Which T-SQL statement should be used to instruct a database management system to use an index instead of performing a full table scan?
    1. A
      WHERE
    2. B
      SELECT
    3. C
      UNION
    Show answer

    Correct answer: A

    Why

    Filtering in the WHERE clause on an indexed column allows the optimizer to perform an index seek instead of a full table scan. SELECT only lists the columns to return, and UNION combines result sets.