Microsoft Certified: Azure Cosmos DB Developer Specialty Intermediate Practice Exam: Medium Difficulty 2025

You are designing a data model for an e-commerce application using Azure Cosmos DB for NoSQL. The application needs to display product information along with the latest 5 customer reviews on the product detail page. Products have hundreds of reviews that grow over time. Which data modeling approach should you implement?

Your application uses Azure Cosmos DB with the Session consistency level. A user updates their profile information through a web application, and immediately after, a background job triggered by the same user session needs to read this updated data. Which approach ensures the background job reads the latest data?

You are implementing a container in Azure Cosmos DB to store IoT telemetry data. Each device sends approximately 100 messages per day, and you have 50,000 devices. The data must be retained for 90 days for compliance purposes, then automatically deleted. Queries primarily filter by deviceId and timestamp. What is the most cost-effective configuration?

Your Azure Cosmos DB container is experiencing throttling errors (429 status codes) during peak hours between 9 AM and 11 AM. The container is configured with 10,000 RU/s manual throughput. Analysis shows that queries against the container are consuming high RUs due to cross-partition queries. What combination of actions should you take to optimize costs while resolving the throttling?

You need to implement a multi-region Azure Cosmos DB account for a global application. The application requires writes in two regions (East US and West Europe) with automatic failover capability. Read operations should be served from the nearest region to minimize latency. Which configuration should you implement?

You are developing an Azure Function that processes orders from an Azure Cosmos DB container using the Change Feed. The function should process changes in near real-time and maintain processing state. The function occasionally experiences failures. How should you implement the Change Feed processor?

Your application stores user documents in Azure Cosmos DB with the following structure: userId (partition key), username, email, preferences (object), and orderHistory (array of order IDs). Users frequently update their preferences but rarely modify orderHistory. The orderHistory array can grow to contain hundreds of items. Which optimization should you implement?

You are designing a document structure for a social media application where users can create posts and other users can like these posts. Each post can receive thousands of likes. You need to display the total like count and determine if the current user has liked the post. What data modeling approach should you use?

Your application uses Azure Cosmos DB and needs to integrate with an Azure Synapse Analytics workspace for complex analytical queries on historical data. The operational workload should not be affected by analytical queries. Which integration approach should you implement?

You have an Azure Cosmos DB container with a partition key that has resulted in a hot partition, where 70% of requests go to a single partition key value. The container stores customer orders with customerId as the partition key, and most orders come from a few large customers. How should you resolve the hot partition issue?