Chapter 5 of 5 · AWS SAP-C03
Operational Excellence and Automation
IaC, policy as code, pipelines, deployments, observability, SLOs, scaling, recovery and runbooks.
- 7topics
- 36 minfast track
- 1diagrams and exhibits
- 20%of the exam
Topics in this chapter
Take them in order, or jump to the one you need.
- 1Infrastructure and Policy as CodeChoose CloudFormation or CDK, enforce rules with Hooks, aspects, Config rules and SCPs, keep secrets out of templates and deploy…5 min
- 2CI/CD Pipelines and Artifact SecurityDesign CodePipeline, CodeBuild and CodeDeploy pipelines that scan and sign artifacts, then promote one build across accounts and…Diagram5 min
- 3Deployment Strategies and Failure RecoveryPick blue/green, canary or rolling deployments with rollback, design health check replacement and add circuit breakers, retries…6 min
- 4Observability and Log AggregationCollect metrics, logs and traces with CloudWatch, X-Ray and OpenTelemetry, watch ECS and EKS with Container Insights, and…5 min
- 5Alerting, SLOs and User ExperienceBuild CloudWatch alarms with SNS and Systems Manager actions, watch users with Synthetics and RUM, and set SLOs with error…5 min
- 6Scaling, Capacity and Automated RunbooksChoose target tracking, step, scheduled or predictive scaling, plan capacity with Service Quotas and AWS Health, and automate…5 min
- 7Evaluating and Monitoring AI WorkloadsValidate model and RAG outputs with Amazon Bedrock Evaluations, and monitor latency, tokens, agents and model drift with…5 min
About this chapter
What this chapter covers
This chapter is 20% of the AWS Certified Solutions Architect Professional SAP-C03 exam. It tests how you build, release and run workloads across many accounts and Regions with as little manual work as possible.
The topics run from infrastructure as code and policy as code, through CI/CD pipelines and deployment strategies, to observability, alerting, service level objectives, scaling, automated recovery and runbooks. Two topics cover AI workloads: evaluating model outputs before release and monitoring models and agents in production.
Questions are long multi-account scenarios. Most have a single answer, and multiple-response questions state the count, such as "Which TWO". Watch for discriminators like LEAST operational overhead, MOST scalable or fastest rollback, and expect plain text questions with no exhibits.
How the topics fit together
Operations is a loop: define everything as code, release it safely, watch it, and let automation respond before people have to.
- Infrastructure and policy as code make every account identical and block bad changes before they deploy.
- Pipelines build, scan, sign and promote one artifact through accounts using assumed roles.
- Deployment strategies limit the blast radius of a bad release and roll it back automatically.
- Observability collects metrics, logs and traces with standard instrumentation and centralizes them.
- Alerting and SLOs turn signals into decisions about user experience and error budgets.
- Scaling, recovery and runbooks act on those signals with safety limits so automation cannot make an incident worse.
When two answers both work, the professional exam usually wants the managed, organization-wide option, such as StackSets with automatic deployment or a delegated administrator, over a script run account by account.
How to use these pages
Each topic page has the same shape, so you can skim for what you need.
- What it is: the idea in plain words.
- How it works: the mechanism, with diagrams and examples.
- What to remember for the exam: the facts to memorize.
- Common mistakes: the classic traps.
- Check yourself: three questions to test your understanding.
Focus on the trigger words in each scenario, such as every new account, roll back automatically, without code changes and noisy alerts. They point straight to the service or pattern.
Ready for more practice?
The full AWS SAP-C03 question bank has 850 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by AWS. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.