DP-420 · Question #159
You plan to create an Azure Cosmos DB for NoSQL account that will have a single write region and three read regions. You need to set the consistency level for the account. The solution must meet the…
The correct answer is D. Strong. To meet requirements for synchronous replication across at least three replicas, ordered reads for transactional batches, and maximized throughput in a multi-region Azure Cosmos DB account, Strong consistency should be selected.
Question
You plan to create an Azure Cosmos DB for NoSQL account that will have a single write region and three read regions. You need to set the consistency level for the account. The solution must meet the following requirements:
- In the write region, writes must replicate synchronously across at
least three replicas.
- In the read regions, reads must see writes in order for transactional
batches.
- Throughput for reads and writes must be maximized.
Which consistency level should you select?
Options
- AConsistent Prefix
- BBounded Staleness
- CEventual
- DStrong
How the community answered
(36 responses)- A8% (3)
- B6% (2)
- C3% (1)
- D83% (30)
Why each option
To meet requirements for synchronous replication across at least three replicas, ordered reads for transactional batches, and maximized throughput in a multi-region Azure Cosmos DB account, Strong consistency should be selected.
Consistent Prefix ensures that reads never see out-of-order writes but does not guarantee the absolute latest write or synchronous quorum replication.
Bounded Staleness allows for a defined maximum staleness, meaning reads may not see the very latest write immediately, which fails the requirement for synchronous replication and immediate ordered reads for transactional batches.
Eventual consistency offers no ordering guarantees for reads and does not provide the immediate consistency required for transactional batches.
Strong consistency ensures that reads always return the most recent committed version of an item. In a multi-region account with a single write region, this means writes are synchronously committed to a quorum of replicas (typically 3) within the write region before acknowledgement, and all read regions reflect this state, fulfilling all stated requirements.
Concept tested: Azure Cosmos DB consistency levels and their properties
Source: https://learn.microsoft.com/azure/cosmos-db/consistency-levels
Topics
Community Discussion
No community discussion yet for this question.