Competing Consumers Pattern
Implement the competing consumers pattern to allow multiple consumers to process messages from a single queue. Scale your processing capacity by adding more consumers.
Competing Consumers Pattern is a free RabbitMQ Messaging & Async Systems lesson on CoddyKit — lesson 1 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the RabbitMQ Messaging & Async Systems learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.
Scaling with Competing Consumers
Welcome! In distributed systems, you often need to process many tasks efficiently. The Competing Consumers pattern is a powerful way to achieve this.
It allows you to scale your message processing capacity by simply adding more workers.
How Competing Consumers Work
Imagine a single queue of tasks. Instead of one worker taking all tasks, multiple workers (consumers) listen to this same queue.
- Each message is delivered to only one of the competing consumers.
- Consumers "compete" to receive the next available message.
- This distributes the workload automatically.
Key Benefits of the Pattern
The Competing Consumers pattern offers several advantages:
- Scalability: Easily increase processing power by adding more consumer instances.
- Reliability: If one consumer fails, others can pick up its share of messages.
- Load Balancing: Messages are spread across available consumers, balancing the workload.
- Decoupling: Producers don't need to know how many consumers there are or where they are.
Producer: Sending Tasks
Let's set up a basic producer that sends messages (tasks) to a queue named task_queue. Each message will be a simple string.
Run this code to send a few messages:
import com.rabbitmq.client.Channel;
import com.rabbitmq.client.Connection;
import com.rabbitmq.client.ConnectionFactory;
public class TaskProducer {
private final static String QUEUE_NAME = "task_queue";
public static void main(String[] argv) throws Exception {
ConnectionFactory factory = new ConnectionFactory();
factory.setHost("localhost"); // Assuming RabbitMQ is local
try (Connection connection = factory.newConnection();
Channel channel = connection.createChannel()) {
// Declare a durable queue
channel.queueDeclare(QUEUE_NAME, true, false, false, null);
for (int i = 0; i < 10; i++) {
String message = "Task " + (i + 1);
channel.basicPublish("", QUEUE_NAME, null, message.getBytes("UTF-8"));
System.out.println(" [x] Sent '" + message + "'");
Thread.sleep(100); // Small delay to visualize
}
}
}
}Consumer 1: Processing Tasks
Now, let's create our first consumer. It will connect to task_queue and start processing messages. Each message will be acknowledged after simulating work.
Run this consumer in a terminal:
import com.rabbitmq.client.Channel;
import com.rabbitmq.client.Connection;
import com.rabbitmq.client.ConnectionFactory;
import com.rabbitmq.client.DeliverCallback;
public class TaskConsumer {
private final static String QUEUE_NAME = "task_queue";
public static void main(String[] argv) throws Exception {
ConnectionFactory factory = new ConnectionFactory();
factory.setHost("localhost");
Connection connection = factory.newConnection();
Channel channel = connection.createChannel();
channel.queueDeclare(QUEUE_NAME, true, false, false, null);
System.out.println(" [*] Consumer 1 waiting for messages.");
// Basic QoS: Prefetch 1 message at a time to ensure fair dispatch
channel.basicQos(1);
DeliverCallback deliverCallback = (consumerTag, delivery) -> {
String message = new String(delivery.getBody(), "UTF-8");
System.out.println(" [C1] Received '" + message + "'");
try {
Thread.sleep(1000); // Simulate work
} finally {
System.out.println(" [C1] Done '" + message + "'");
channel.basicAck(delivery.getEnvelope().getDeliveryTag(), false);
}
};
channel.basicConsume(QUEUE_NAME, false, deliverCallback, consumerTag -> {});
}
}Running Multiple Consumers
To truly see the competing consumers pattern in action, open a new terminal window and run the exact same TaskConsumer code again.
You'll now have two consumer instances listening to the task_queue. Run the producer code (from Scene 4) once more. Observe how messages are now distributed between both consumer instances, demonstrating how they compete for messages and share the workload!
RabbitMQ's Dispatching Logic
By default, RabbitMQ uses a round-robin dispatching mechanism when multiple consumers are connected to the same queue. This means messages are sent to consumers sequentially.
- Consumer 1 gets message 1.
- Consumer 2 gets message 2.
- Consumer 1 gets message 3, and so on.
The basicQos(1) setting in our consumer code is crucial here. It tells RabbitMQ not to send more than one unacknowledged message to a consumer at a time, ensuring fair dispatch even if consumers process at different speeds.
Common Use Cases
The Competing Consumers pattern is ideal for scenarios like:
- Image processing: Multiple workers resizing images from a queue.
- Email sending: Sending bulk emails without overwhelming a single service.
- Log processing: Analyzing large volumes of logs in parallel.
- Background jobs: Any task that can be processed independently by multiple workers.
Important Considerations
When using competing consumers, keep these in mind:
- Message Ordering: If strict message order is critical, this pattern might not be suitable directly, as messages can be processed out of order by different consumers.
- Idempotency: Consumers should ideally be idempotent. This means processing the same message multiple times should have the same effect as processing it once. This is vital for fault tolerance and retries.
Competing Consumers Quiz
Test your understanding of the Competing Consumers pattern.
Recap: Competing Consumers
Great job! You've learned about the Competing Consumers pattern:
- It enables multiple consumers to process messages from a single queue.
- It's excellent for scaling and load balancing message processing.
- RabbitMQ's default round-robin dispatch and QoS settings facilitate fair distribution.
- Consider idempotency and potential out-of-order processing for specific use cases.
Next, we'll dive deeper into optimizing consumer efficiency with prefetch counts!
Frequently asked questions
Is the “Competing Consumers Pattern” lesson free?
Yes — the full text of “Competing Consumers Pattern” is free to read here on the web, and the RabbitMQ Messaging & Async Systems course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the RabbitMQ Messaging & Async Systems course, upgrade to CoddyKit PRO.
What will I learn in “Competing Consumers Pattern”?
Implement the competing consumers pattern to allow multiple consumers to process messages from a single queue. Scale your processing capacity by adding more consumers. You practise RabbitMQ Messaging & Async Systems with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.
Do I need any experience to start RabbitMQ Messaging & Async Systems?
No prior experience is required. RabbitMQ Messaging & Async Systems on CoddyKit is structured for beginners through advanced learners; this is — lesson 1 of 4, so you can start here or from the beginning and move at your own pace.
How long does the “Competing Consumers Pattern” lesson take?
Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.
Can I write and run code in this RabbitMQ Messaging & Async Systems lesson?
Yes. Every RabbitMQ Messaging & Async Systems lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.
All lessons in this course
- Competing Consumers Pattern
- Prefetch Count (QoS)
- Exclusive Consumers & Consumer Priority
- Single Active Consumer