Hard1 markMultiple Choice

AZ-305 · Question 28 · Domain 2.3: Recommend a Data Storage Solution

You are designing an Azure Cosmos DB database for a global IoT application.

Millions of devices will write telemetry data continuously. The data will be queried heavily by a dashboard that filters by DeviceId and Date. You need to select a partition key that ensures even distribution of storage and throughput (Request Units) to avoid 'hot partitions'.

Which TWO partition key strategies would be effective? (Select TWO)

Answer options:

A.

Date

B.

A synthetic key combining DeviceId and Date

C.

DeviceId

D.

DeviceType

E.

TenantId (assuming all devices belong to one tenant)

How to approach this question

Look for high cardinality (many unique values) and even write distribution. Avoid properties that group all current writes together (like Date).

Full Answer

In Cosmos DB, the partition key determines how data is distributed across physical partitions. `DeviceId` is a good choice because with millions of devices, it has high cardinality and spreads writes evenly. A synthetic key combining `DeviceId` and `Date` is also excellent, as it prevents any single device's partition from growing too large over time while still supporting the dashboard's query pattern.

Common mistakes

Selecting 'Date'. Partitioning by date is an anti-pattern for write-heavy workloads because all writes for the current day hit a single physical partition, causing throttling.

Practice the full Azure Solutions Architect Expert AZ-305 Practice Exam 4

55 questions · hints · full answers · grading

More questions from this exam