Describe considerations for working with non-relational data on Azure →mediumMultiple ChoiceObjective-mapped
DP-900 Practice Question: Describe considerations for working with non-relational data on Azure
A mobile app stores user preferences in Azure Cosmos DB using the NoSQL API. The app frequently reads a single user's profile by user ID (the partition key). The development team wants the fastest possible read performance globally and is willing to accept that reads might not reflect the latest write immediately. Which consistency level should they choose to minimize read latency?
⚠ Common exam trap
Many exam-takers confuse 'fastest read performance' with 'strongest consistency' and choose Strong or Bounded staleness, not realizing that the question explicitly allows stale reads, making Eventual the optimal choice for minimizing latency.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Eventual
Eventual consistency offers the lowest read latency because it allows reads to return data from any replica without waiting for confirmation that the write has been fully replicated. Since the app can tolerate stale reads (i.e., not reflecting the latest write immediately), Eventual consistency eliminates the synchronization overhead required by stronger models, making it the fastest choice for global read performance.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Strong
Why it's wrong here
Strong consistency in Cosmos DB ensures reads always return the most recent committed write, but this requires synchronously replicating to a quorum of replicas before acknowledging the write, increasing latency and reducing availability during regional failures. For a mobile app storing user preferences, this trade-off is unnecessary; users can tolerate slightly stale profile data. The key point: strong consistency is overkill for this scenario, so it's not the optimal choice.
- ✗
Bounded staleness
Why it's wrong here
Bounded staleness is a consistency level that allows reads to lag behind writes by at most a defined time interval (e.g., 5 seconds) or a maximum number of versions, whichever comes first. It still requires replicas to track a lag boundary, causing more overhead and higher read latency than eventual consistency for globally distributed data. While it would work for user preferences, it doesn't give the lowest possible latency that eventual provides, making it a worse fit for this read-heavy, latency-sensitive mobile app scenario.
- ✗
Session
Why it's wrong here
Session consistency scopes read-your-writes and monotonic reads to a single client session, using a session token to synchronize replicas for that session. This adds complexity and overhead because the client must manage session tokens, and global reads outside the session may still be stale; also, the write path may have extra coordination to maintain the session's guarantees. For a mobile app where the user is the one viewing their preferences, the session-level guarantee is more than needed—eventual will also eventually show the write, and the app can accept that tiny delay without the extra session overhead.
- ✓
Eventual
Why this is correct
Eventual consistency is the default in Cosmos DB for multi-region writes and offers the lowest latency and highest availability because replicas converge asynchronously without waiting for quorum. For mobile user preferences, the profile is typically read frequently and written occasionally; if a user updates a preference and reads it a moment later, eventual might show the old value briefly, but it will converge quickly. Because user preferences are non-critical and tolerate a brief stale read, eventual provides the optimal balance of performance and cost, making it the correct choice.
Go deeper
Related to this question
Learn chapter
Data Roles and Core Concepts
Key term
Azure Cosmos DB
Azure Cosmos DB is a fully managed, globally distributed NoSQL database service that offers fast reads and writes anywhere in the world with automatic scaling and multiple consistency models.
Key term
Data
Data is raw, unprocessed information, like numbers, words, or measurements, that can be stored, processed, and analyzed by computers.
About these practice questions
Courseiva writes every DP-900 question from scratch — 820 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This DP-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the DP-900 exam.