What message brokers have you worked with? Tell me about the difference between Kafka and RabbitMQ.

I worked with Kafka and RabbitMQ — two popular message brokers, but with different architectures and purposes.

Kafka:

• Built as a distributed message log.
• Messages are stored on disk and accessible to multiple consumers.
• Supports high throughput and scalability.
• Ideal for stream processing, analytics, and event handling.
• Consumers manage their own offsets, providing flexibility in reading.

RabbitMQ:

• Based on message queues supporting various routing models (exchange, binding).
• Messages are delivered and removed from the queue after acknowledgment.
• Supports complex routing scenarios and delivery guarantees.
• Suitable for systems requiring reliable delivery and complex routing logic.

Main differences:

• Kafka is oriented towards storing and processing large data streams with re-read capability.
• RabbitMQ is a classic queue broker for message passing with delivery guarantees.

Example usage:

// Example of sending a message to RabbitMQ
Channel channel = connection.createChannel();
channel.queueDeclare("task_queue", true, false, false, null);
String message = "Hello World";
channel.basicPublish("", "task_queue", MessageProperties.PERSISTENT_TEXT_PLAIN, message.getBytes());

In Kafka, sending and receiving messages is done through Producer and Consumer APIs with offset management.