In a monolithic application, we usually use one database for everything. This single database handles both read and write operations. That means that the database is both working for complex join queries, and also perform CUD (Create, Update, Delete) operations. But, when the application gets more complicated, handling these read and write tasks can become really hard to manage.

CQRS stands for Command Query Responsibility Segregation. It's a design pattern that splits the operations of a system into two distinct parts: commands (which modify data i.e., writes) and queries (which read data i.e., reads). This separation allows for more efficient handling and scaling of each operation.

Why Use CQRS in Microservices?

1. Scalability: Separate scaling of read and write workloads. You can scale your query and command services independently based on demand.
2. Optimized Performance: Each service can be optimized for its specific task (read or write), leading to improved performance.
3. Simplified Complexity: By separating concerns, it becomes easier to manage the complexity of each microservice.
4. Flexibility in Technology Choice: Different technology stacks can be used for the read side and the write side if it suits the specific needs better.