Design PatternsEventdriven architecture
Event-driven architecture is a style of design pattern in which components of a system communicate with each other by exchanging events. An event can be anything from a user action (such as clicking a button) to a system-generated notification (such as an error message). This communication typically happens asynchronously, meaning that the sending and receiving of events is not directly tied to each other in time.
Event-driven architecture has several benefits over traditional synchronous communication paradigms. First, it allows for greater scalability since events can be processed in parallel by multiple components. Second, it decouples event producers from event consumers, which makes it easier to change or add new components without affecting existing ones. Finally, Event-driven architecture often leads to more resilient systems since individual components can continue functioning even if other parts of the system are unavailable.
Despite these advantages, Event-driven architecture does have some drawbacks. First, it can be difficult to debug since errors may not surface until long after the original event was generated. Second, certain types of events may need to be handled in a specific order, which can be tough to ensure with an asynchronous system. Nonetheless, Event-driven architecture is a powerful tool that can be used to build scalable, flexible, and robust systems.
How Pub/Sub works
In pub/sub systems, components subscribe to events they are interested in and receive events when they are published. Pub/sub systems are usually asynchronous, which means that event consumers don't have to be active when an event is published in order to receive the event. Message brokers are often used to implement pub/sub systems. Some popular message brokers are Apache Kafka and RabbitMQ.
How Observer works
In observer systems, components register interest in specific events with an observable object. When the observable object raises an event, all registered observers are notified synchronously. The observers then handle the event as they see fit. The Java platform has built-in support for observer patterns through the java.util.Observable class and the java.util.Observer interface.
EDA systems can be implemented using either type of architecture, but pub/sub systems are more common because they're more scalable and easier to maintain than observer systems.