Azure Data Fundamentals (DP-900) Exam Preparation: Analytics workloads

1. Question 1 of 20

   In a large-scale analytics solution, what happens in the data ingestion and processing layer?

   • ASource data is loaded into a lake or warehouse with ETL or ELT
   • BEach sale is committed to the database with ACID guarantees
   • CBusiness users view finished dashboards on their phones
   • DDatabase administrators grant and revoke user permissions
   Show answer

   Answer: A. Source data is loaded into a lake or warehouse with ETL or ELT

   The ingestion layer brings data from transactional stores, files and streams into an analytical store, usually with ETL or ELT.

2. Question 2 of 20

   In a star schema, which table holds the numeric values you want to analyze, such as sales amounts?

   • AA view
   • BThe fact table
   • CA dimension table
   • DA lookup index
   Show answer

   Answer: B. The fact table

   Fact tables hold the numeric measures. Dimension tables hold the entities you aggregate by, such as customer, product, store and time.

3. Question 3 of 20

   A data warehouse links a Product dimension table to a separate Category table. What is this design called?

   • AA snowflake schema
   • BA key-value schema
   • CA lambda architecture
   • DA star schema
   Show answer

   Answer: A. A snowflake schema

   A snowflake schema adds more tables to the dimensions to represent hierarchies, such as products grouped into categories.

4. Question 4 of 20

   Data lakes often apply a tabular schema to files only when the data is read for analysis. What is this approach called?

   • AReferential integrity
   • BSchema-on-read
   • CSchema-on-write
   • DNormalization
   Show answer

   Answer: B. Schema-on-read

   Data lakes use schema-on-read, so files are stored without constraints and a schema is applied when the data is queried.

5. Question 5 of 20

   Which open-source format adds relational table features on top of Parquet files and is the standard lakehouse format for Microsoft Fabric and Azure Databricks?

   • AXML
   • BCSV
   • CDelta Lake
   • DAvro
   Show answer

   Answer: C. Delta Lake

   Delta Lake builds on Parquet by adding a transaction log, which enables ACID transactions, versioning and schema enforcement.

6. Question 6 of 20

   Which statement describes Microsoft Fabric?

   • AA relational database engine based on SQL Server
   • BA file-sharing service that uses the SMB and NFS protocols
   • CA unified SaaS analytics platform built on OneLake
   • DA key-value store for fast lookups by mobile apps
   Show answer

   Answer: C. A unified SaaS analytics platform built on OneLake

   Fabric brings data engineering, data warehousing, real-time analytics, data science and Power BI together in one SaaS workspace on OneLake.

7. Question 7 of 20

   Which platform is built on Apache Spark, runs as a managed service inside your Azure subscription and uses Delta Lake as its native format?

   • AAzure Databricks
   • BAzure Table storage
   • CAzure Files
   • DAzure SQL Database
   Show answer

   Answer: A. Azure Databricks

   Azure Databricks is a cloud analytics platform built on Apache Spark for data engineering, data science and SQL analytics.

8. Question 8 of 20

   Your data is structured, your team writes SQL, and you need strong schema enforcement for many users at once. Which Fabric item fits best?

   • ADataflows Gen2
   • BEventstream
   • CFabric Lakehouse
   • DFabric Warehouse
   Show answer

   Answer: D. Fabric Warehouse

   Fabric Warehouse is a fully managed, SQL Server-compatible relational warehouse that suits structured data and strong schema enforcement.

9. Question 9 of 20

   A team wants an Azure SQL Database to be copied into OneLake continuously, in near real time, without building pipelines. Which Fabric feature fits?

   • AShortcuts
   • BDataflows Gen2
   • CMirroring
   • DReal-Time Dashboards
   Show answer

   Answer: C. Mirroring

   Mirroring continuously replicates external databases such as Azure SQL Database or Azure Cosmos DB into OneLake in Delta Lake format.

10. Question 10 of 20

    What is a OneLake shortcut?

    • AA saved KQL query that runs in an Eventhouse
    • BA scheduled pipeline that copies data every night
    • CA pre-aggregated cube of monthly sales totals
    • DA live reference to external data, with no copying
    Show answer

    Answer: D. A live reference to external data, with no copying

    A shortcut points to data where it already lives, such as ADLS Gen2 or Amazon S3. There is no pipeline, movement or duplication.

11. Question 11 of 20

    A credit card company sends each customer one bill a month for all of that month's purchases. Which kind of processing is this an example of?

    • AReal-time alerting
    • BTransactional processing
    • CBatch processing
    • DStream processing
    Show answer

    Answer: C. Batch processing

    In batch processing, records are collected and stored, then processed together as a group, such as a monthly bill.

12. Question 12 of 20

    Which statement correctly compares stream processing with batch processing?

    • AStream latency is usually seconds or less; batch latency is usually hours
    • BBatch processing gives results as soon as each new event arrives
    • CStream processing can always use the whole dataset, while batch processing sees only recent data
    • DStream processing is better for complex analytics over very large datasets
    Show answer

    Answer: A. Stream latency is usually seconds or less; batch latency is usually hours

    Stream processing works on each event as it arrives, with latency in seconds or milliseconds. Batch latency is typically a few hours.

13. Question 13 of 20

    A streaming query calculates the average temperature over the last 30 seconds, again and again. What is the query using?

    • AA snowflake schema
    • BA full scan of all historical data
    • CA rolling time window
    • DA composite primary key
    Show answer

    Answer: C. A rolling time window

    Stream processing usually sees only recent data or a rolling time window, and it suits simple calculations such as rolling averages.

14. Question 14 of 20

    Which service is a data ingestion service similar to Azure Event Hubs but optimized for event data from Internet of things (IoT) devices?

    • AAzure Database for MySQL
    • BAzure Stream Analytics
    • CAzure Files
    • DAzure IoT Hub
    Show answer

    Answer: D. Azure IoT Hub

    Azure IoT Hub is like Event Hubs, but it is optimized to manage event data from IoT devices.

15. Question 15 of 20

    Which Azure service is a PaaS option for streaming jobs that read from a source, apply a perpetual query and write the results to an output?

    • AAzure Stream Analytics
    • BAzure Data Lake Storage Gen2
    • CAzure SQL Managed Instance
    • DPower BI Desktop
    Show answer

    Answer: A. Azure Stream Analytics

    Azure Stream Analytics runs streaming jobs with a perpetual query, and it suits standalone or hybrid streaming outside Fabric.

16. Question 16 of 20

    In Microsoft Fabric Real-Time Intelligence, an Eventhouse stores time-series and event data. Which query language do you use with it?

    • APL/SQL
    • BMQL (MongoDB Query Language)
    • CCQL (Cassandra Query Language)
    • DKQL (Kusto Query Language)
    Show answer

    Answer: D. KQL (Kusto Query Language)

    An Eventhouse is optimized for time-series and event data and is queried with KQL, a language built for fast log and telemetry analysis.

17. Question 17 of 20

    Which part of Fabric Real-Time Intelligence triggers automated actions when streaming data meets conditions you set?

    • AEventhouse
    • BActivator
    • CReal-Time hub
    • DOneLake file explorer
    Show answer

    Answer: B. Activator

    Activator turns insights into actions by firing alerts or actions when the data meets defined conditions.

18. Question 18 of 20

    What is the usual Power BI workflow?

    • AWrite the report in Azure Data Factory, then publish it to Azure Files
    • BBuild reports on the phone app, then copy them into Power BI Desktop
    • CBuild in Power BI Desktop, publish to the Power BI service, then share
    • DCreate the report in the service, then download it to Desktop to share it
    Show answer

    Answer: C. Build in Power BI Desktop, publish to the Power BI service, then share

    Analysts build in Power BI Desktop, publish to the Power BI service for sharing and scheduled refresh, and users consume reports in a browser or on the phone app.

19. Question 19 of 20

    In a Power BI semantic model, users view sales by year and then drill down to months and days. What makes this possible?

    • AA rolling time window
    • BA hierarchy in the time dimension
    • CA foreign key in the fact table
    • DA page blob
    Show answer

    Answer: B. A hierarchy in the time dimension

    Hierarchies let you drill up or down through aggregated values at different levels of a dimension, such as year, month and day.

20. Question 20 of 20

    A report must show how marketing spend relates to revenue for each campaign, to see if there is a correlation. Which visual fits best?

    • AA pie chart
    • BA map
    • CA scatter plot
    • DA card
    Show answer

    Answer: C. A scatter plot

    Scatter plots compare two numeric measures and help you see a relationship or correlation between them.

Keyboard: 1–4 to answer, Enter for the next question.