How do the characteristics of working with databases differ when using a microservices architecture?

In a microservices architecture, each service typically owns its own database (Database per Service), which differs from the monolithic approach with a single shared database. This provides service independence but leads to the following features:

• Data isolation: Each microservice is responsible for its own database, reducing conflict risks and simplifying scaling.
• Lack of transactions between services: Traditional distributed transactions (ACID) are difficult to implement, so eventual consistency and saga patterns are often used.
• Different types of databases: Microservices can use different DBMSs optimized for their specific tasks (SQL, NoSQL, etc.).
• Integration complexity: To obtain aggregated data from multiple services, API calls or event-driven approaches are used.

Example: The order service stores its data in one database, the user service in another. To get complete order information with user data, the order service calls the user service via API, not directly to its database.