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 global social media app uses Azure Cosmos DB (NoSQL API) to store user profile data. The app is read-heavy and must serve content with the lowest possible latency to users worldwide. The data is updated by users, and the business has determined that eventual consistency is acceptable because immediate consistency after a write is not critical for profile views. Which consistency level should they choose to minimize read latency?
⚠ Common exam trap
Watch out — candidates often choose Session consistency because it is the default for many Azure Cosmos DB SDKs, but the question explicitly asks for the lowest read latency with eventual consistency acceptable, making Eventual the correct answer despite Session being a common default.
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 is the correct choice because it offers the lowest read latency by allowing reads to return data immediately without waiting for replication to complete. Since the app is read-heavy, global, and can tolerate eventual consistency for profile views, this consistency level minimizes the time to serve content by not imposing any ordering or staleness guarantees on replicas.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Eventual
Why this is correct
In Azure Cosmos DB NoSQL API, Eventual consistency is the weakest consistency level and the default; updates replicate asynchronously with no ordering guarantee, so reads can be served from any replica with minimal latency and maximum availability. A user viewing a social feed may briefly see missing or old posts, but replication converges once writes stop, and read performance is not constrained by quorum acknowledgments.
- ✗
Strong
Why it's wrong here
Strong consistency guarantees linearizable reads by confirming the latest committed write across a quorum of replicas before returning the result. That quorum coordination adds measurable round-trip latency per read and creates availability vulnerability in a globally distributed social media app; it also cannot be configured when multiple regions accept writes simultaneously, which sharply limits the write scalability a global timeline requires.
- ✗
Bounded staleness
Why it's wrong here
Bounded staleness lets you specify a maximum lag window (e.g., 5 seconds or 100 operations) and guarantees reads stay within that window, but that guarantee is not free: reads may need to wait for or verify enough replicas to prove the staleness bound is satisfied. It also requires maintaining an ordered history and cannot be combined with multi-region writes, so for a social media app that explicitly needs absolute minimal read latency, its added coordination overhead makes it slower than Eventual.
- ✗
Session
Why it's wrong here
Session consistency scopes guarantees to a single client session, providing read-your-writes and monotonic reads for that session by passing a session token with every request. A global social media app receives requests from countless anonymous, stateless, or cross-device clients, so maintaining session tokens adds protocol overhead and read latency because Cosmos DB must validate the token against replica state; that per-session coordination is inherently more expensive than the uncoordinated reads of Eventual.
Go deeper
Related to this question
Learn chapter
Data Roles and Core Concepts
Key term
Consistency level
Consistency level is a setting in Azure data services that determines how quickly and accurately data is synchronized across multiple copies in a distributed system.
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.
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.