What were the bottlenecks in the monolith messenger, and why was it decomposed?

In a monolithic messenger, bottlenecks often included:

• Scalability: all functions (sending messages, storage, authentication) operate within a single application, making it difficult to scale individual parts under load.
• Response time: with a large number of users, the monolith can slow down due to locks and shared state.
• Development and deployment: changes in one part require rebuilding and redeploying the entire application, slowing down releases.
• Fault tolerance: a failure in one component can affect the entire system.

Decomposition allowed for the separation of individual services (e.g., message service, user service, notification service), which improved scalability, simplified maintenance, and accelerated development. Each service can be scaled and updated independently, increasing the system's fault tolerance.