3.5 Management and Governance Services
Key Takeaways
- AWS CloudFormation is Infrastructure as Code (IaC) — define AWS resources in YAML or JSON templates and deploy them consistently.
- AWS Systems Manager provides a unified interface for managing AWS resources at scale, including patch management and parameter store.
- AWS CloudWatch monitors AWS resources and applications, collecting metrics, setting alarms, and providing dashboards.
- AWS Trusted Advisor provides real-time guidance across five categories: cost optimization, performance, security, fault tolerance, and service limits.
- AWS Health Dashboard shows the status of AWS services that affect your specific resources and account.
Quick Answer: AWS management and governance tools let you define infrastructure as code (CloudFormation/CDK), monitor resources (CloudWatch), audit activity (CloudTrail), get recommendations (Trusted Advisor), manage at scale (Systems Manager), and track service status (AWS Health). The CLF-C02 leans hard on distinguishing CloudWatch from CloudTrail and knowing Trusted Advisor's five categories.
Infrastructure as Code (IaC)
AWS CloudFormation
AWS CloudFormation models and provisions AWS resources from declarative templates written in JSON or YAML.
Key benefits:
- Repeatable deployments — the same template builds identical environments every time
- Version control — store templates in Git to track infrastructure changes
- Automatic rollback — if creation fails, CloudFormation rolls the stack back
- Free — you pay only for the resources a template creates
- Stacks — a collection of resources managed as one unit; delete the stack to remove everything
Example: define a full web app (VPC, EC2, RDS, S3, CloudFront) in one template, deploy it to dev/staging/prod identically, then tear it all down by deleting the stack.
AWS CDK (Cloud Development Kit)
The AWS CDK lets you define infrastructure using real programming languages (TypeScript, Python, Java, C#, Go); it synthesizes CloudFormation templates under the hood — ideal for developers who prefer code over YAML.
Monitoring and Observability
Amazon CloudWatch
Amazon CloudWatch is the primary monitoring and observability service.
| Feature | Description |
|---|---|
| Metrics | Collect CPU, network, disk, and custom metrics from AWS services |
| Alarms | Trigger actions on thresholds (e.g., scale out when CPU > 70%) |
| Logs | Centralize, search, and analyze application/system logs |
| Dashboards | Build visualizations of metrics and alarms |
| EventBridge | React to state changes across AWS resources |
Common uses: drive Auto Scaling from CPU utilization, set billing alarms to catch runaway spend, centralize logs for troubleshooting, and build operational dashboards.
On the Exam: CloudWatch = monitoring (metrics, alarms, logs, dashboards — "how are my resources performing?"). CloudTrail = auditing (who made which API call, when, from where). Do not confuse them; this exact pairing appears repeatedly.
AWS X-Ray
AWS X-Ray traces and debugs distributed applications and microservices, giving an end-to-end view of a request as it crosses services to pinpoint bottlenecks and errors.
Advisory and Optimization
AWS Trusted Advisor
Trusted Advisor gives real-time recommendations across five categories:
| Category | What It Checks |
|---|---|
| Cost Optimization | Idle/underutilized instances, unattached EBS volumes |
| Performance | Overutilized resources, high-utilization instances |
| Security | Open security groups, MFA on root, exposed access keys |
| Fault Tolerance | Missing Multi-AZ, missing backups |
| Service Limits | Resources approaching service quota limits |
| Support Tier | Trusted Advisor Checks |
|---|---|
| Basic & Developer | Core checks (security and service-limits focused) |
| Business, Enterprise On-Ramp & Enterprise | All checks + API access + CloudWatch integration |
On the Exam: Memorize the five categories (a useful mnemonic: C-P-S-F-S). Full Trusted Advisor requires Business support or higher.
AWS Compute Optimizer
Uses machine learning to recommend optimal compute resources (EC2 instances, Auto Scaling groups, EBS volumes, Lambda functions, ECS on Fargate) based on actual utilization.
Account and Resource Management
AWS Systems Manager
Systems Manager is a unified interface for operating resources at scale:
| Feature | Description |
|---|---|
| Session Manager | Secure shell to EC2 with no SSH keys or bastion hosts |
| Patch Manager | Automate OS and software patching |
| Parameter Store | Centralized config and secrets storage |
| Run Command | Execute commands across a fleet of instances |
| Inventory | Collect metadata about managed instances |
AWS Config
AWS Config records resource configurations over time and evaluates them against rules for compliance — answering "what did this resource look like last Tuesday, and is it compliant?" It complements CloudTrail (which records the API calls themselves).
AWS Health Dashboard
| View | What It Shows |
|---|---|
| Service health | General AWS service status across Regions |
| Your account health | Events affecting your specific account and resources |
On the Exam: The account-specific (personal) view of AWS Health shows events affecting your resources; the service view shows general AWS status.
Deployment and Developer Tools
| Service | Purpose |
|---|---|
| AWS CodeCommit | Managed Git repos (closed to new customers in 2024 — use GitHub/GitLab) |
| AWS CodeBuild | Managed build service (compile, test, package) |
| AWS CodeDeploy | Automated deployment to EC2, Lambda, ECS |
| AWS CodePipeline | CI/CD pipeline orchestration tying the stages together |
| AWS CloudShell | Browser-based shell with the AWS CLI pre-configured |
Quick Reference
| Need | Service |
|---|---|
| Infrastructure as Code | CloudFormation or CDK |
| Monitor metrics and set alarms | CloudWatch |
| Audit API activity | CloudTrail |
| Track config history/compliance | AWS Config |
| Optimization recommendations | Trusted Advisor |
| Patch and operate fleets | Systems Manager |
| CI/CD pipeline | CodePipeline |
| Debug microservices | X-Ray |
Governing Multiple Accounts
Larger organizations rarely run everything in one account. AWS Organizations groups many accounts under one management account, enabling consolidated billing, centralized policy, and grouping accounts into organizational units. Service Control Policies (SCPs) then set permission guardrails across those accounts — they cap what any identity in an account can do, even an administrator, regardless of individual IAM permissions.
For new multi-account setups, AWS Control Tower automates a secure, well-architected landing zone using Organizations, SCPs, and best-practice configurations out of the box. On the exam, "enforce guardrails across many accounts" points to SCPs in Organizations, and "set up a governed multi-account environment quickly" points to Control Tower.
Which AWS service allows you to define your cloud infrastructure using JSON or YAML templates?
AWS Trusted Advisor provides recommendations in which FIVE categories?
Which service should you use to set an alarm when your EC2 CPU utilization exceeds 80%?