Explore / Microsoft Azure Data Fundamentals (DP-900) Exam Prep / Practice questions
Microsoft Azure Data Fundamentals (DP-900) practice questions
10 questions from the bank behind this course's timed mock. Free to read and to answer, no account needed, and each has its answer and an explanation. They belong to Microsoft Azure Data Fundamentals (DP-900) Exam Prep.
Answer them one at a time →Read lesson 1 free
Question 1
A data analyst failed their first attempt at Exam DP-900 this morning. They want to book a retake as soon as Microsoft's retake policy allows. When is the earliest they can retake the exam?
- Immediately, on the same day
- 24 hours after the first attempt
- 14 days after the first attempt
- After they complete the free Practice Assessment
Show the answer
Answer: B. 24 hours after the first attempt
Microsoft's policy allows a retake 24 hours after a first failed attempt. Longer waits only apply to later retakes, and finishing the Practice Assessment does not change when you can rebook.
Question 2
A data engineer at a manufacturing company receives the following file from an older supplier system: <Invoice number="5521" currency="GBP"> <Line sku="B7" qty="4"/> </Invoice> Which file format is it?
- XML
- CSV
- Avro
- JSON
Show the answer
Answer: A. XML
The tags in angle brackets < > that define elements and attributes show that this is XML. JSON is the nearest distractor, but JSON puts objects in braces { } and writes attributes as name:value pairs.
Question 3
A bank's auditor reviews a completed transfer of $40 from Account A (balance $100) to Account B (balance $50). The auditor confirms that Account A now shows $60, Account B shows $90, and the two accounts together still hold $150. Which ACID property does this check confirm?
- Isolation
- Consistency
- Durability
- Atomicity
Show the answer
Answer: B. Consistency
Consistency means a transaction moves the database from one valid state to another, so the balances show the transfer and the total is unchanged. Atomicity is the nearest distractor, but it is about the debit and credit both happening or neither happening, not about the end state being valid.
Question 4
Users report that customer lookups in a company's live order-processing database have become slow during business hours. The company wants the database to run consistently fast again. Which role is mainly responsible for investigating and tuning it?
- Data analyst
- Data scientist
- Data engineer
- Database administrator
Show the answer
Answer: D. Database administrator
Keeping an operational database available and performing well, and tuning it when it slows down, is a core DBA responsibility. A data engineer builds and monitors pipelines that move data out of the database, but does not own the live database's performance.
Question 5
A product manager asks a database developer to record a supplier for every product. The existing Product table has no Supplier column, and the existing rows and data must be kept. Which type of SQL statement should the developer use?
- A DDL statement such as DROP TABLE followed by CREATE TABLE
- A DDL statement such as ALTER TABLE
- A DCL statement such as GRANT
- A DML statement such as UPDATE
Show the answer
Answer: B. A DDL statement such as ALTER TABLE
Adding a column changes the table's structure, so it needs the DDL statement ALTER TABLE, and the existing rows stay in place. UPDATE is DML and can only change values in columns that already exist. DROP TABLE would permanently delete every row.
Question 6
A finance team runs an application on Azure Database for PostgreSQL Flexible Server. Auditors require the team to restore the database to any point in time within the last 30 days. The server still uses the default backup settings. What should the database administrator do?
- Increase the backup retention period on the server to at least 30 days
- Do nothing, because the default retention already covers 30 days
- Enable zone-redundant high availability to extend the restore window
- Migrate the database to SQL Server on Azure Virtual Machines to get longer backups
Show the answer
Answer: A. Increase the backup retention period on the server to at least 30 days
Flexible Server keeps backups for 7 days by default and can be set to keep them for up to 35 days, so the administrator only needs to raise the retention period. Zone-redundant high availability adds a standby server for failover and does not change how long backups are kept.
Question 7
A retail analyst creates a new general-purpose v2 storage account to hold daily sales exports. She does not choose an access tier for the account or for any blob. Which tier do newly uploaded block blobs use?
- Cool
- Hot
- Cold
- Archive
Show the answer
Answer: B. Hot
New general-purpose v2 accounts use hot as their default access tier, so blobs uploaded without a tier are stored as hot. Archive can never be an account's default tier, and cool or cold apply only if someone sets them.
Question 8
A retail developer is building a new product catalogue from scratch. A shoe has sizes and colours, and a TV has screen size and ports. The developer wants to store each item as JSON without empty columns, and the app has no ties to an existing database. Which option should the developer choose?
- Azure Table storage
- Azure SQL Database
- Azure Cosmos DB for NoSQL
- Azure Cosmos DB for Apache Cassandra
Show the answer
Answer: C. Azure Cosmos DB for NoSQL
The API for NoSQL is the native document (JSON) API and is the recommended starting point for a new document app with no legacy constraint. Azure Cosmos DB for Apache Cassandra uses a column-family model and is meant for moving existing CQL apps.
Question 9
A healthcare analytics team must remove patient identifiers and fix invalid date formats in source records. Compliance rules say no uncleaned data may ever be stored in the analytical store. Which data processing approach meets this constraint?
- Copying the raw records into a lakehouse and cleaning them with Apache Spark
- ETL (extract, transform, load)
- ELT (extract, load, transform)
- Schema-on-read in a data lake
Show the answer
Answer: B. ETL (extract, transform, load)
ETL transforms data before it is loaded, so only cleaned data reaches the analytical store. ELT loads the raw copy into the store first and transforms it afterwards, which breaks the compliance rule.
Question 10
An energy company streams smart-meter readings to a live operations dashboard. The same readings are also saved and analysed later, together with nightly batch-loaded billing data in a data warehouse. What is this common pattern called?
- Lambda architecture
- Perpetual query
- Micro batch processing
- Rolling time window
Show the answer
Answer: A. Lambda architecture
A lambda architecture sends streamed data to a real-time view and also saves it for historical analysis next to batch-processed data in an analytical store. Micro batch processing only describes how a stream engine handles a few records at a time. It does not describe combining real-time and historical paths.
The full timed mock
50 questions in 45 minutes, with a pass mark of 70%. Each paper is drawn from a bank of questions written for the mock alone, and these 10 are left out of it. Full access opens it, with every level of the course.
LevelUpWith is independent of the organisation that sets this exam. These questions were written for this course; they are not taken from the exam.