Event-Driven Architecture & CQRS

CQRS separates the write path that enforces rules from the read path that serves fast, purpose-built views, so each side is modeled and scaled for its own job.

CQRS in plain Spring Boot is commands as records, a rule-enforcing JPA aggregate that publishes events, a transactional event listener that updates a read table, and separate command and query controllers.

CQRS buys independent scaling and fast tailored reads at the price of eventual consistency, duplicated data, and more moving parts, so skip it for simple, low-traffic, strongly-consistent CRUD.

Domain events are private, in-process, and may use your internal model; integration events are public, broker-carried, versioned contracts that shield consumers from your internals and are keyed by aggregate ID for ordering.

Choreography is services reacting to each other's events with no central brain; orchestration is a central coordinator explicitly driving each step and owning the workflow.

Pinball architecture is a complex workflow bouncing through services with no coordinator, which destroys observability and compensation; the fix is an orchestrated Saga with a central state machine.

Event Notification sends a tiny pointer and makes consumers call back; Event-Carried State Transfer sends the data consumers need so they act independently; never send the whole aggregate just in case.

Event Sourcing stores every change as an immutable event and rebuilds current state by replaying the log, so the history is the truth and the balance is just a calculation.