Professional Cloud DevOps Engineer study guide: exam domains, plan and common mistakes
Written and reviewed by the DevOpsCode Lab Team. Last updated 19 November 2026.
Professional Cloud DevOps Engineer is Google Cloud’s certification for engineers who build and run delivery pipelines and production services. It checks that you can bootstrap a governed Google Cloud environment, build CI/CD pipelines for applications, infrastructure and ML workloads, apply site reliability engineering practices, run observability and troubleshooting, and keep performance and cost in balance.
What the exam tests
Questions are scenarios from a team’s working week: a release that broke an SLO, a pipeline that needs a security gate, an alert that fires too often, a bill that grew too fast. You choose the service, setting or practice that fixes it with the least effort and risk. SRE thinking from Google’s own practice, such as error budgets, blameless postmortems and reducing toil, sits next to product knowledge of Cloud Build, Cloud Deploy, Artifact Registry and Google Cloud Observability.
Domains and how much each counts
| Domain | Weight | What to be able to do | Questions in our bank |
|---|---|---|---|
| Establishing and bootstrapping foundational infrastructure and organizational governance | 18% | Design the resource hierarchy and projects for teams and environments, manage infrastructure as code with Terraform or Infrastructure Manager, set up shared networking and access controls | 43 |
| Building, implementing, and maintaining CI/CD pipelines for application, infrastructure, and ML workloads | 30% | Build pipelines with Cloud Build, store artifacts in Artifact Registry, release with Cloud Deploy, apply canary and blue/green strategies, secure the supply chain with Binary Authorization and Secret Manager, deliver ML models with Vertex AI | 72 |
| Applying site reliability engineering practices | 18% | Define SLIs and SLOs, manage error budgets, reduce toil, run incident response and blameless postmortems, plan capacity | 43 |
| Implementing observability practices and troubleshooting issues | 22% | Use Cloud Monitoring, Cloud Logging, Cloud Trace and Managed Service for Prometheus; build log sinks, log-based metrics, dashboards and alerting policies; debug production issues | 53 |
| Optimizing performance and cost | 12% | Right-size resources with recommenders, use committed use discounts and Spot VMs, tune autoscaling, analyse billing exports | 29 |
CI/CD and observability together are more than half of the exam.
Exam format
Our practice tests use 50 questions in 120 minutes with a 70% pass mark. The real exam is two hours of multiple choice and multiple select questions with no hands-on lab, and can be taken online or at a test centre. Check the official Google Cloud exam page for the current format, price and objectives before you book.
Who should take it
- DevOps, platform and SRE engineers who run services on Google Cloud.
- Release and build engineers who own CI/CD pipelines.
- Associate Cloud Engineer holders who work on delivery and operations.
Google suggests several years of industry experience, including at least one year on Google Cloud.
A 4-week study plan
| Week | Focus | Practice |
|---|---|---|
| 1 | Foundations and SRE: resource hierarchy for environments, Terraform, SLIs, SLOs, error budgets, toil, postmortems | Domain quizzes for both; one full test at the end of the week as a baseline |
| 2 | CI/CD: Cloud Build triggers and steps, Artifact Registry, Cloud Deploy targets and promotion, canary and blue/green, Binary Authorization, Secret Manager | Domain quiz; build a small pipeline from a Git repository to Cloud Run in a trial project |
| 3 | Observability and cost: metrics, logs, traces, log sinks and exclusions, alerting on SLO burn rate, recommenders, discounts | Two domain quizzes; review every wrong answer with its explanation |
| 4 | Full exams and weak spots | Two full timed tests, then the mistakes quiz; book the exam when you score 80%+ twice |
Topics people often get wrong
- SLO versus SLA and what an error budget does. The SLO is the internal reliability target; the error budget is the allowed unreliability under it. When the budget runs out, the usual policy is to slow or freeze feature releases and focus on reliability, not to change the SLO.
- Alerting on SLOs. Alert on burn rate of the error budget rather than on every individual error or CPU spike. Fast burn needs a page; slow burn can be a ticket.
- Log routing. The Log Router sends logs to destinations through sinks, and each sink has its
own exclusion filters. Excluding logs from the
_Defaultsink stops them being stored in the_Defaultbucket, but other sinks still route them. Retention is set per log bucket. - Canary versus blue/green. A canary sends a small share of traffic to the new version and increases it if metrics stay healthy. Blue/green runs two full environments and switches all traffic at once, which makes rollback fast but needs double capacity.
- Where policy enforcement belongs. Binary Authorization blocks deployment of images that lack the required attestations. Vulnerability scanning in Artifact Registry finds problems, but it does not block a deployment by itself.
How to use the practice tests
Start with the free Professional Cloud DevOps Engineer questions to see the style. Then take one full timed test in exam mode for a baseline and look at the score by domain. Spend most of your time in the two weakest domains with domain quizzes, and finish with the mistakes quiz before each new full test. The readiness number tells you when your recent scores hold above the pass mark across all domains.
Frequently asked questions
How hard is the Professional Cloud DevOps Engineer exam? It is a professional exam that mixes product knowledge with SRE judgement. Engineers who run Google Cloud services often need 4–6 weeks of focused practice.
Do I need to read the Google SRE books? They help. Many questions on SLOs, error budgets, toil and postmortems follow the practices described in Google’s free SRE books.
Do I need Associate Cloud Engineer first? No. There is no prerequisite, but ACE covers the IAM, networking and compute basics that this exam assumes.
Are these the real exam questions? No. Every question is written by us for the published exam guide. We never use leaked or recalled exam content.
DevOpsCode Lab is not affiliated with or endorsed by Google Cloud. Exam names are trademarks of their owners. Check the official Google Cloud exam page for the current price, format and objectives before you book.