Message broker

As explained in Platform Engineering on Kubernetes

A message broker is an intermediary service that receives, routes, and delivers messages between application components.

A message broker receives, routes, and delivers messages between application components. In asynchronous messaging, it decouples producers from consumers: a consumer can connect later, and services can isolate message flows through topics. section 7.1.4

Using a message broker introduces several implementation concerns:

• Producers and consumers must know the broker location and how to publish or subscribe. section 7.1.4
• Topics group messages so services sharing one broker can exchange different message streams. section 7.1.4
• Applications require broker-specific client libraries, protocols, versions, credentials, and connection configuration. section 7.1.4

Kafka in a Kubernetes application

A Conference application runs Kafka alongside its services, Redis, and PostgreSQL in a Kubernetes cluster. The services use Kafka to exchange asynchronous messages, while pod readiness and restart behavior reveal whether dependencies such as Kafka are available. section 2.2.1

Figure 7.10 Asynchronous messaging interactions

Use the interaction diagram to trace a producer placing a message into the broker and a later consumer retrieving it. section 7.1.4

Figure 7.11 Dependencies and API challenges

Use the dependency diagram to inspect how Kafka client libraries and broker-specific APIs affect application services. section 7.1.4

How Platform Engineering on Kubernetes covers it

How deep it goes

Explains producer-consumer timing, topics, dependencies, protocols, client libraries, and configuration.

Examples

Uses Kafka in a Conference application deployed with Kubernetes, Redis, PostgreSQL, and multiple application services.

What it assumes

Assumes service-to-service communication knowledge and contrasts asynchronous messaging with HTTP and gRPC.

What you can do after

Apply it via Kafka to decouple services, isolate message groups with topics, and account for client dependencies and connection settings.

How it connects

Connects asynchronous messaging with HTTP, gRPC, topics, CloudEvents, Kafka clients, and Kubernetes services.

liveBookmanning.com

Read Platform Engineering on Kubernetes on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Data Pipelines with Apache Airflow, Second EditionA broker is a message-queue component that receives, stores, and delivers messages between producers and consumers.
• 

  liveBookmanning.com
  Spring in Action, Sixth EditionA message broker is middleware that accepts, routes, and delivers messages between application components or services.

Compare how 3 books explain message broker →

Related concepts

• RabbitMQincludes · RabbitMQ is an open-source message broker that routes and delivers messages between producers and consumers using queues and messaging protocols.
• unmanaged dependencykind of · An unmanaged dependency is an out-of-process dependency whose interactions are observable by external systems and that is not fully controlled by the application.
• Apache Kafkaincludes · Apache Kafka is a distributed event-streaming platform for publishing, storing, processing, and consuming streams of records.
• messageuses · A message is a discrete unit of information or data represented, processed, or exchanged in a communication context.
• event-driven architectureused by · Event-driven architecture is a software architectural style in which components communicate or initiate processing by producing and responding to events.

This explainer was generated by Manning's knowledge graph from the text of Platform Engineering on Kubernetes. The section links point to the parts of the book it draws on.