How is information exchanged between microservices in a system architecture?

Data exchange between microservices is usually implemented via two main approaches:

1. Synchronous interaction — using REST API or gRPC. Microservices make HTTP requests to each other, waiting for responses. This is simple and understandable but can lead to delays and resilience issues.

2. Asynchronous interaction — via message exchange systems (message brokers), such as Kafka, RabbitMQ, or ActiveMQ. Microservices publish events or messages to queues, and others subscribe to them. This approach enhances scalability and system resilience.

Example of synchronous REST call in Java with Spring Boot:

@RestController
public class OrderController {

    private final PaymentService paymentService;

    public OrderController(PaymentService paymentService) {
        this.paymentService = paymentService;
    }

    @PostMapping("/order")
    public ResponseEntity<String> createOrder(@RequestBody Order order) {
        boolean paymentResult = paymentService.processPayment(order.getPaymentInfo());
        if (paymentResult) {
            return ResponseEntity.ok("Order created");
        } else {
            return ResponseEntity.status(HttpStatus.PAYMENT_REQUIRED).body("Payment failed");
        }
    }
}

Here, PaymentService can be a client of another microservice.

For asynchronous exchange, Kafka Producer and Consumer are used, where one service publishes an event, and another listens and processes it.