AWS CDK

Exam relevance: Domain 2 (29%). CDK lets you define infrastructure in familiar programming languages. Know constructs, stacks, apps, and how CDK relates to CloudFormation.

1. Overview

AWS Cloud Development Kit (CDK) is an open-source framework for defining cloud infrastructure using familiar programming languages (TypeScript, Python, Java, C#, Go). CDK synthesizes to CloudFormation templates.

CDK vs CloudFormation:

2. Core Concepts

App

• Root of the CDK application
• Contains one or more stacks

Stack

• Unit of deployment — maps to a CloudFormation stack
• Contains constructs (resources)
• Can be deployed to a specific account and region

Constructs

Three levels of abstraction:

// L2 construct — sensible defaults, type-safe
const bucket = new s3.Bucket(this, 'MyBucket', {
  versioned: true,
  encryption: s3.BucketEncryption.KMS_MANAGED,
  removalPolicy: cdk.RemovalPolicy.RETAIN,
});

// L3 pattern — entire ALB + Fargate service in one construct
const service = new ecs_patterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  cpu: 512,
  taskImageOptions: { image: ecs.ContainerImage.fromRegistry('nginx') },
});

3. CDK Workflow

# Initialize a new CDK app
cdk init app --language typescript

# Synthesize CloudFormation template
cdk synth

# Compare deployed stack with current state
cdk diff

# Deploy stack
cdk deploy

# Destroy stack
cdk destroy

4. CDK Constructs Library

• AWS Constructs Library: L2 and L3 constructs for all AWS services
• Construct Hub: community and third-party constructs
• Share constructs as npm/PyPI packages across teams

5. CDK Pipelines

Self-mutating CI/CD pipeline for CDK applications:

Source (CodeCommit/GitHub)
  └── CDK Pipeline (CodePipeline)
        ├── Synth stage (cdk synth)
        ├── Self-mutation (update pipeline itself)
        └── Deploy stages (dev → staging → prod)

• Pipeline updates itself when pipeline code changes
• Supports multiple stages with approval gates
• Cross-account and cross-region deployments

6. Exam Scenarios

Scenario 1: A development team wants to define AWS infrastructure using Python with IDE autocomplete and type checking, and reuse infrastructure patterns across multiple projects. What is the solution?

Use AWS CDK with Python. Define infrastructure using L2/L3 constructs. Package reusable patterns as Python libraries published to a private PyPI repository. CDK synthesizes to CloudFormation for deployment.

Scenario 2: A company wants a CI/CD pipeline that automatically deploys CDK infrastructure changes through dev, staging, and prod environments with manual approval before prod. What is the solution?

Use CDK Pipelines. Define a self-mutating pipeline with stages for each environment. Add a manual approval step before the prod stage. The pipeline updates itself when pipeline code changes, and deploys application stacks when application code changes.

7. Exam Tips

• CDK synthesizes to CloudFormation — all CloudFormation features apply
• L3 constructs (patterns) are the most powerful — entire architectures in a few lines
• CDK Pipelines = self-mutating — the pipeline updates itself when you change pipeline code
• CDK is not a replacement for CloudFormation — it generates CloudFormation templates
• Use cdk diff before deploying to preview changes (equivalent to CloudFormation change sets)
• CDK apps can deploy to multiple accounts and regions in a single cdk deploy

8. Official References

• CDK Developer Guide
• CDK API Reference
• CDK Pipelines
• Construct Hub