A weather service stores sensor readings with fields: stationId, date, hour, and temperature. The write rate is very high, and queries always filter by stationId and a date range. The current partition key is stationId, but some stations generate far more data than others, causing hot partitions. Which approach best solves the hot-partition problem while keeping queries efficient?

Use stationId as the partition key and increase throughput.

Use a synthetic partition key that combines stationId and date.

A container uses a partition key that has only three distinct values. What is the likely impact on data distribution across physical partitions?

Data will be concentrated in a few logical partitions, potentially causing hot partitions

Data will be stored in a single physical partition regardless of the partition key

Data distribution will be optimal because the partition key has low cardinality

Data will be evenly distributed across all physical partitions automatically

A banking app stores account documents and transaction documents. A stored procedure must atomically transfer funds between two accounts: it reads both account documents and writes updated balances. The partition key for both containers is accountId. Which design ensures the transfer is a single transaction?

Run the stored procedure on each account's partition separately.

Store both accounts in the same partition by using a synthetic partition key that groups the two accounts.

Use a client-side transaction that spans both partitions.

Use a change feed to reconcile balances after the transfer.

An order management system stores orders and order items in the same container. Each order has a unique orderId and a customerId. A stored procedure must atomically update an order and all its items. Which partition key enables this transaction?

Use a synthetic key combining customerId and orderId.

Use a separate container for order items.

Use customerId as the partition key.

Use orderId as the partition key.

In Azure Cosmos DB for NoSQL, how does the choice of partition key affect throughput distribution?

Throughput is provisioned per logical partition, so each partition key value gets its own RU allocation

Throughput is provisioned at the container level and is shared across all physical partitions, but uneven data or access patterns can cause throttling on hot partitions

Throughput is only available on the partition key that is most frequently accessed

Throughput is automatically scaled based on the number of distinct partition key values

Microsoft Certified:Azure Cosmos DB Developer Specialty

Domain 1Objective 2

Design a Data Partitioning Strategy for Azure Cosmos DB for NoSQL DP-420 Practice Questions (Page 5)

Part of the Design and implement data models domain, which makes up ~45% of our current practice bank. Microsoft does not publish an official question count, but from its 100-minute exam (~40–65 total, ~18–29 in this domain), expect 3–4 from this objective — we provide 28 practice questions to prepare you well beyond it. (estimate)

28questions here

6free pages

9concepts

Questions 21–25

21application · hard
A weather service stores sensor readings with fields: stationId, date, hour, and temperature. The write rate is very high, and queries always filter by stationId and a date range. The current partition key is stationId, but some stations generate far more data than others, causing hot partitions. Which approach best solves the hot-partition problem while keeping queries efficient?
Select an answer first
22foundation · easy
A container uses a partition key that has only three distinct values. What is the likely impact on data distribution across physical partitions?
Select an answer first
23application · hard
A banking app stores account documents and transaction documents. A stored procedure must atomically transfer funds between two accounts: it reads both account documents and writes updated balances. The partition key for both containers is accountId. Which design ensures the transfer is a single transaction?
Select an answer first
24application · medium
An order management system stores orders and order items in the same container. Each order has a unique orderId and a customerId. A stored procedure must atomically update an order and all its items. Which partition key enables this transaction?
Select an answer first
25foundation · easy
In Azure Cosmos DB for NoSQL, how does the choice of partition key affect throughput distribution?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to DP-420

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “DP-420” is a trademark of its owner, used for identification only.

Design a Data Partitioning Strategy for Azure Cosmos DB for NoSQL DP-420 Practice Questions (Page 5)

Questions 21–25

A container uses a partition key that has only three distinct values. What is the likely impact on data distribution across physical partitions?

An order management system stores orders and order items in the same container. Each order has a unique orderId and a customerId. A stored procedure must atomically update an order and all its items. Which partition key enables this transaction?

In Azure Cosmos DB for NoSQL, how does the choice of partition key affect throughput distribution?

Keep studying — Design and implement data models