DP-900 Describe Considerations for Non-Relational Data Practice Questions

Free DP-900 Describe Considerations for Non-Relational Data practice test questions

8 questions from this domain with answers and explanations - different from the samples on the main DP-900 page. A free account includes a fixed set of 20 questions for this exam; paid access unlocks the full bank.

1. Question 1Describe Considerations for Non-Relational Data

   A social media company needs to store user profiles as JSON documents. Each profile can have different attributes depending on the user type (personal, business, creator). The application must support global distribution with single-digit millisecond read latency. Which Azure service is the best choice?

   • ATable Storage
   • BAzure Cosmos DBCorrect
   • CBlob Storage
   • DAzure Files
   ✓ Correct answer: B

   Azure Cosmos DB is a globally distributed NoSQL database that stores JSON documents with automatic multi-region replication. It guarantees single-digit millisecond read latency at global scale and supports flexible schemas where different documents can have different attributes. For the social media scenario with varying user profile structures and global distribution requirements, Cosmos DB provides the performance, flexibility, and scalability needed.

   Why the other options are wrong

   • ATable Storage is incorrect because it stores simple key-value entities and does not offer global distribution with single-digit millisecond reads or rich JSON document queries.
   • CBlob Storage is incorrect because it is object storage for large files, not a queryable JSON document store distributed worldwide.
   • DAzure Files is incorrect because it provides SMB file shares that mount as drives, not a schema-flexible document database with millisecond global reads.
2. Question 2Describe Considerations for Non-Relational Data

   A web application stores user profile photos in Azure Blob Storage. The operations team wants to organize the photos into a two-level hierarchy, with a top-level grouping per year and virtual folders by month beneath it. In Azure Blob Storage, what is the highest-level organizational unit that directly holds blobs?

   • AStorage account
   • BBlob tag
   • CSnapshot
   • DContainerCorrect
   ✓ Correct answer: D

   In Azure Blob Storage the hierarchy is storage account -> container -> blob. A container is the highest-level unit that directly holds blobs; every blob lives inside exactly one container, and a storage account can contain many containers. Blob Storage has no true nested folders, so a year/month structure is created with virtual folders - slashes in the blob name that tools display as a hierarchy - rather than real directories. The container therefore sits directly above the blobs, with virtual folders only a naming convention beneath it.

   Why the other options are wrong

   • AA storage account is the parent resource that contains containers, but it does not hold blobs directly; the container is the immediate holder of blob objects, and virtual folder names such as "year/month/photo.jpg" only look like a hierarchy inside it.
   • BA blob tag is a key-value attribute applied to an individual blob for filtering and search, not an organizational unit that holds blobs.
   • CA snapshot is a read-only point-in-time copy of a single blob kept for versioning, not a unit that holds blobs.
3. Question 3Describe Considerations for Non-Relational Data

   Wingtip Toys needs to store product catalog data where each product can have a completely different set of attributes. For example, a toy might have age range and material, while a book might have author and page count. Which type of NoSQL data store is best suited for this scenario?

   • AA columnar database
   • BA document databaseCorrect
   • CA graph database
   • DA key-value store
   ✓ Correct answer: B

   Document databases excel at storing entities with varying schemas and attributes, making them perfect for product catalogs where different product types have completely different properties. Unlike relational databases that enforce schemas, document stores like MongoDB and Cosmos DB allow each document to have its own structure, enabling flexible attribute sets.

   Why the other options are wrong

   • AA columnar database is incorrect because columnar stores require predefined column schemas and are optimized for analytical queries on many rows, not flexible schema scenarios.
   • CA graph database is incorrect because graph databases optimize relationship traversal, not flexible attribute storage.
   • DA key-value store is incorrect because key-value stores do not support complex queries or filtering by different attributes within values.
4. Question 4Describe Considerations for Non-Relational Data

   Lucerne Publishing needs a simple key-value store in Azure to cache frequently accessed session data for their web application. Which Azure service is most appropriate?

   • AAzure Synapse Analytics
   • BAzure Data Lake Storage
   • CAzure Cache for RedisCorrect
   • DAzure SQL Database
   ✓ Correct answer: C

   Azure Cache for Redis is a fully managed, in-memory data store specifically designed for caching frequently accessed data with extremely low latency access patterns. It provides a key-value store model where you can store and retrieve session data using simple key lookups, making it ideal for web application session caching. Redis supports various data structures (strings, hashes, lists, sets, sorted sets) and offers sub-millisecond response times, making it perfect for high-performance caching scenarios where data needs to be accessed with minimal latency.

   Why the other options are wrong

   • AAzure Synapse Analytics is incorrect because it is a data warehouse and analytics service designed for complex analytical queries on large datasets, not for caching simple session data.
   • BAzure Data Lake Storage is incorrect because it is an object storage service for big data analytics, not optimized for fast key-value lookups or caching frequently accessed session data.
   • DAzure SQL Database is incorrect because it is a relational database optimized for transactional workloads and complex queries, not designed for the low-latency, high-throughput requirements of session caching.
5. Question 5Describe Considerations for Non-Relational Data

   In Azure Cosmos DB, a partition key determines how data is distributed across physical partitions for scalability.

   • ATrueCorrect
   • BFalse
   ✓ Correct answer: A

   A container's data is split into logical partitions by partition key value, and each logical partition is placed on a physical partition holding a bounded amount of storage and throughput. A single logical partition never spans physical partitions, which is what makes the key choice consequential: if one key value attracts most of the writes, that logical partition becomes a hot partition whose throughput is capped no matter how much is provisioned for the container overall. A key with high cardinality and even access spreads both storage and request units across physical partitions. The key is also the efficient query filter, since a query that supplies it is routed to one partition instead of fanning out to all of them.

   Why the other options are wrong

   • BIn Azure Cosmos DB, the partition key is a critical design element that determines how data is logically and physically distributed across partitions. The partition key value is hashed to create a partition ID, which maps the data to specific physical partitions on backend servers.
6. Question 6Describe Considerations for Non-Relational Data

   Azure Cosmos DB guarantees single-digit millisecond read and write latency at the 99th percentile when deployed with the proper consistency level.

   • ATrueCorrect
   • BFalse
   ✓ Correct answer: A

   The guarantee is stated at the 99th percentile because the interesting property is the tail rather than the average, and it covers point reads and writes by id and partition key rather than arbitrary queries. The configured consistency level is part of the bargain: strong consistency must hear from a quorum of replicas before acknowledging, which raises write latency, and that is why the low-latency figure is quoted for the weaker levels such as session or eventual. Meeting it also depends on the request being served from a region near the client and on throughput being provisioned high enough that requests are not rate-limited, since a throttled request is retried rather than served.

   Why the other options are wrong

   • BAzure Cosmos DB is specifically architected to deliver single-digit millisecond latency for both read and write operations at the 99th percentile of requests. This latency guarantee is a core feature of the platform when configured with appropriate consistency levels such as eventual or session consistency.
7. Question 7Describe Considerations for Non-Relational Data

   A data engineer must move a 200 GB on-premises backup file into Azure Blob Storage as quickly and reliably as possible from a command line. Which tool is purpose-built for high-performance bulk uploads of blobs?

   • AAzCopyCorrect
   • Bbcp
   • CSSMS
   • DSSRS
   ✓ Correct answer: A

   AzCopy is a Microsoft command-line utility optimized for fast, resilient bulk transfer of blobs and files between local storage and Azure. It supports parallelism, resumable transfers, and large files, which makes it ideal for uploading a large backup file quickly and reliably.

   Why the other options are wrong

   • Bbcp is a bulk copy utility that moves rows between SQL Server tables and flat files, not a tool for uploading blobs to a storage account.
   • CSQL Server Management Studio administers and queries SQL databases; it performs no blob transfers to storage accounts.
   • DSQL Server Reporting Services renders paginated reports and performs no bulk uploads to Azure Blob Storage.
8. Question 8Describe Considerations for Non-Relational Data

   Contoso wants to copy data written to a container in Azure Cosmos DB for NoSQL into an Azure Synapse Analytics environment for analytics, without affecting the performance of their transactional workload. Which Azure Cosmos DB capability is designed for this?

   • AAzure Synapse Link with the analytical storeCorrect
   • BMulti-region write (multi-master) across paired regions
   • CAutoscale provisioned throughput on the container
   • DSession consistency on the client connection
   ✓ Correct answer: A

   Synapse Link automatically syncs data from the transactional (row-based) store into a separate column-based analytical store, which Azure Synapse Analytics can query. Because analytics run against the analytical store, the transactional workload's request units are not consumed, achieving hybrid transactional/analytical processing (HTAP) with no ETL.

   Why the other options are wrong

   • BMulti-region writes let several regions accept writes for lower latency and higher availability, but every replica is still the transactional store with no analytics-optimized copy.
   • CAutoscale raises and lowers provisioned RU/s to match demand; analytics queries would still hit the transactional store and compete for the same throughput.
   • DSession consistency is a read-consistency guarantee for one client session and does nothing to move data into Azure Synapse Analytics.

How Describe Considerations for Non-Relational Data is tested

This domain holds 136 of the 709 questions in the DP-900 bank, about 19%. The mix is 101 single-answer multiple choice, 16 multiple-response, 15 true/false, 2 yes/no scenario, 2 ordering and 1 hotspot, so it is worth practising the formats as well as the content.

Once you have a few attempts recorded, CertGrid scores every domain separately and points you at the weakest one, so you can drill Describe Considerations for Non-Relational Data on its own rather than re-running full-length mocks.

Other DP-900 exam domains

• Describe Core Data Concepts195 questions
• Identify Considerations for Relational Data174 questions
• Describe an Analytics Workload204 questions
• All DP-900 practice questions709 total
• Describe Considerations for Non-Relational Data study notesKey concepts
• Microsoft practice examsAll Microsoft

DP-900 Describe Considerations for Non-Relational Data FAQ