Question 1

A logistics firm stores 12,000 shipment records as JSON documents. Most records carry a trackingId and destination, but about 30% also include an optional customsNotes array. Which term best describes this data?

Accepted Answer

Semi-structured data JSON documents are self-describing: tags and keys carry the structure, yet individual entities may vary, for example some have customsNotes and others don't. That is semi-structured data. option A needs one fixed schema for every row, option B has no organizing markers at all (for example images or free text), and option C describes a storage format for raw bytes rather than the shape of the data.

Question 2

A bank debits 300 USD from account A and credits 300 USD to account B in one transaction. The database must ensure that either both changes are saved or neither is. Which ACID property is being described?

Accepted Answer

Atomicity Atomicity treats a transaction as an indivisible unit that fully succeeds or is fully rolled back. option A keeps the data valid against rules such as constraints, option C keeps concurrent transactions from seeing each other's partial changes, and option D ensures committed data survives a crash.

Question 3

Managers need a reusable object that presents the result of a saved query joining 3 tables as if it were a table, without storing a separate copy of the data. Which database object should be created?

Accepted Answer

A view A view is a virtual table defined by a query; it stores the query, not the data. A clustered index (option A) defines physical row order, a stored procedure that inserts rows (option B) modifies data, and a heap (option D) is a table without a clustered index.

Question 4

A blob in the Archive tier must be read again by an analyst. What must happen first?

Accepted Answer

The blob must be rehydrated to the Hot or Cool tier Archive blobs are offline, so they must first be rehydrated to an online tier such as Hot or Cool, which can take hours. Copying the blob within the Archive tier (option B) keeps it offline, converting it to a page blob (option C) is not possible, and a shared access signature (option D) grants access but cannot make offline data readable.

Question 5

Microsoft Fabric stores all workloads' data in a single, tenant-wide logical data lake based on Delta Parquet. What is this called?

Accepted Answer

OneLake OneLake is the unified, tenant-wide data lake of Microsoft Fabric, often described as the OneDrive for data. The other names are tools or features for streaming, alerts and reports.

Question 6

A company needs a scalable service to ingest 1 million telemetry events per second from applications before processing them. Which Azure service is built for this?

Accepted Answer

Azure Event Hubs Event Hubs is a big data streaming platform built to ingest millions of events per second. Queue storage (option A) handles messages for asynchronous work but not high-throughput streaming, Managed Instance (option B) is a relational database, and Stream Analytics (option C) processes streams but relies on an ingestion service such as Event Hubs.

Free Azure Data Fundamentals (DP-900) questions

Want the full DP-900 question bank?

What these free questions cover

Lila Wolke Creator