Application Integration Services

Why this matters

A modern application is rarely one program. It is many small parts that hand work to each other: a web layer takes an order, another component charges the card, a third updates inventory, and a fourth sends an email. If those parts call each other directly, one slow or failed component can freeze the whole system.

Application integration services solve this. They let the parts of an application talk to each other without being tightly connected, which is called decoupling. The components pass messages or events through a managed service instead of calling each other directly, so one part can fail, slow down, or scale up on its own without breaking the rest. The CLF-C02 exam tests 4 of these services at a recognition level: Amazon SQS, Amazon SNS, Amazon EventBridge, and AWS Step Functions. Your job is to match a described need to the right one.

Amazon SQS: message queues

Amazon Simple Queue Service (Amazon SQS) is a fully managed message queue. A producer puts messages into the queue, and a consumer pulls them out to process one at a time, at its own pace. The queue stores the messages in between, so the producer and consumer never have to be available at the same moment.

This is the classic way to decouple a fast front end from slower back-end work. Imagine a website that accepts photo uploads. Instead of making the user wait while each photo is resized, the site drops a message in an SQS queue and responds right away. A separate worker reads the queue and resizes the photos when it can. If the worker falls behind or restarts, the messages wait safely in the queue.

SQS offers 2 queue types you should recognize by name:

• Standard queues give maximum throughput and at-least-once delivery, with best-effort ordering.
• FIFO queues (first-in, first-out) guarantee that messages are processed exactly once, in the exact order they were sent.

Think of SQS as point-to-point: each message is meant for one consumer to handle.

Amazon SNS: pub/sub fan-out

Amazon Simple Notification Service (Amazon SNS) is a managed publish/subscribe (pub/sub) messaging service. Publishers send messages to a topic, and the topic pushes a copy of each message to every subscriber. This is many-to-many, or fan-out: one message reaches many destinations at once.

Subscribers can be SQS queues, AWS Lambda functions, HTTP/S endpoints, and even end users through SMS, email, and mobile push. So SNS is the service to reach for when a single event needs to trigger several reactions in parallel, or when you need to send notifications to people.

A very common design is SNS plus SQS together. The publisher sends one message to an SNS topic, and the topic delivers it to several SQS queues. Each queue feeds a different consumer, so billing, shipping, and analytics can all react to the same order event without knowing about each other.

The fastest way to tell SNS and SQS apart: SQS is a queue that one consumer pulls from, while SNS pushes each message to many subscribers.

Amazon EventBridge: the event bus

Amazon EventBridge is a serverless event bus. An event bus receives a stream of events and routes each one to the right place based on rules you define. A rule matches events by their content, then sends matching events to one or more targets, such as a Lambda function, an SQS queue, or Step Functions.

What sets EventBridge apart is where the events come from. It can receive events from your own applications, from many AWS services, and from third-party SaaS applications like Zendesk or Shopify. That makes it the natural choice for building event-driven systems that react to activity across tools you do not control.

A simple way to hold the distinction: SNS broadcasts a message to its subscribers, while EventBridge inspects each event and routes it by rules, including events from SaaS partners.

AWS Step Functions: orchestrating workflows

The first 3 services move individual messages or events. AWS Step Functions is different: it coordinates a whole sequence of steps as a single workflow.

You define the steps as a visual workflow (a state machine), and Step Functions runs them in order, passes data from one step to the next, retries steps that fail, and tracks the state of the entire process. This is orchestration. It fits any multi-step job where order and error handling matter, such as processing an order, running a data pipeline, or approving a request.

Picture an order workflow: validate the payment, then reserve inventory, then send a confirmation. If reserving inventory fails, Step Functions can retry it or branch to a different path, all without you writing the plumbing to track where the process is.

Putting them side by side

Exam tips

• Application integration services exist to decouple the parts of an application so they can fail and scale independently.
• Amazon SQS is a message queue. One consumer pulls each message. Standard queues favor throughput, FIFO queues guarantee order and exactly-once processing.
• Amazon SNS is pub/sub. It fans out one message to many subscribers (SQS queues, Lambda, HTTP, SMS, email, push). SNS plus SQS is the classic fan-out pattern.
• Tell SQS and SNS apart: SQS is a queue one consumer reads, SNS pushes to many subscribers at once.
• Amazon EventBridge is a serverless event bus that routes events by rules and can ingest events from SaaS applications.
• AWS Step Functions orchestrates multi-step workflows with built-in ordering, retries, and error handling. Pick it when a scenario describes coordinating several steps.