Event-Driven Architecture in Achieving Evolutionary Systems

Applications nowadays are expected to be global, cloud-native, superelastic, and have 99.999% uptime.

As the world keeps changing, new issues arise. That’s where event-driven architecture (EDA) comes in handy. This allows for repeated evaluation and processing and provides a “time machine”.

Event-Driven Architecture also offers horizontal scaling and a shared language with the business.

In this new approach, we discard traditional messaging and send events without API coupling to a remote service. This enables processing events as reactions without the emitter calling on a specific function. It’s a unique approach that simplifies the architecture and allows for greater flexibility.

In the past, transactional consistency was standard practice before NoSQL.

With NoSQL came the concept of eventual consistency through the CAP theorem. Even with eventual consistency, losing data is not an option. Failure is unacceptable. Since the system and events can evolve, we must use log and log-aware stream processors to rebuild and restore the application state.

With an event-driven architecture, we can achieve evolutionary architecture.