G O O G L E C E R T I F I C A T I O N
Professional Cloud Developer Practice Exam
Exam Number: 1006 | Last updated April 21, 2026 | 1246+ questions across 5 vendor-aligned objectives
The Professional Cloud Developer exam measures whether candidates can the skills needed to work on Google Cloud in production settings. The target audience includes application developers who build, deploy, and operate cloud-native workloads on Google Cloud runtimes. Passing candidates have shown they can reason about trade-offs and pick the right service for a given constraint.
Heavy-weighted areas define where study time pays back fastest: 25% targets Designing Highly Scalable, Available, and Reliable Cloud-Native Applications (event-driven patterns, idempotency, autoscaling, graceful degradation); 20% targets Building and Testing Applications (Cloud Build, Artifact Registry, unit and integration testing, Cloud Code); 20% targets Deploying Applications (Cloud Run, Google Kubernetes Engine, App Engine, blue-green and canary patterns).
Supporting domains fill out the blueprint: 20% covers Integrating Google Cloud Services (Pub/Sub, Cloud Tasks, Eventarc, Cloud Endpoints, secrets management); 15% covers Managing Application Performance Monitoring (Cloud Trace, Cloud Profiler, Cloud Logging, error budgets). Each still appears on the exam, so none can be safely skipped.
Every answer links to the source. Each explanation below includes a hyperlink to the exact Google documentation page the question was derived from. PowerKram is the only practice platform with source-verified explanations. Learn about our methodology →
743
practice exam users
96.5%
satisfied users
85.8%
passed the exam
4.7/5
quality rating
Test your Cloud Developer knowledge
10 of 1246+ questions
Question #1 - Designing Highly Scalable, Available, and Reliable Cloud-Native Applications
A payments service receives duplicate HTTP POSTs from flaky mobile clients on retry, causing double charges.
Which application pattern most directly prevents duplicate side effects?
A) Store client-supplied idempotency keys and deduplicate on the server
B) Double the request timeout
C) Increase container memory
D) Disable HTTPS to reduce retries
Show solution
Correct answers: A – Explanation:
Server-side idempotency keys let a retried request map to the same result, which prevents double side effects. Timeouts, memory, and disabling HTTPS do not address duplicates at the semantic level. Source: Check Source
Question #2 - Building and Testing Applications
A team wants every commit on the main branch to run unit tests, build a container image, and push it to Artifact Registry automatically.
Which Google Cloud service provides that CI pipeline?
A) Cloud Monitoring
B) Cloud Scheduler
C) Cloud Build with a cloudbuild.yaml
D) Cloud Tasks
Show solution
Correct answers: C – Explanation:
Cloud Build runs pipelines defined in cloudbuild.yaml, including tests, image builds, and pushes to Artifact Registry. Monitoring is observability. Scheduler triggers cron jobs. Cloud Tasks is an async queue, not CI. Source: Check Source
Question #3 - Deploying Applications
A developer’s service is HTTP-only, stateless, and traffic is bursty with long idle periods overnight.
Which runtime is the simplest fit that scales to zero?
A) GKE Standard with a node pool reserved 24×7
B) Cloud Run
C) Compute Engine VM with a fixed size
D) App Engine Flex with min instances of 5
Show solution
Correct answers: B – Explanation:
Cloud Run scales to zero for bursty idle workloads and is the simplest managed option. A reserved GKE pool keeps nodes warm. A fixed VM does not scale to zero. App Engine Flex with min instances of 5 never scales to zero. Source: Check Source
Question #4 - Integrating Google Cloud Services
A developer needs to decouple producers from consumers so many services can fan out on a single order event.
Which Google Cloud service is the right messaging backbone?
A) Cloud Storage object change notifications alone
B) Cloud Scheduler periodic polling
C) Memorystore Redis pub/sub without subscriptions
D) Pub/Sub topics with multiple subscriptions
Show solution
Correct answers: D – Explanation:
Pub/Sub with multiple subscriptions provides durable fan-out across independent consumers, which is the stated requirement. Object change notifications are storage-specific. Scheduler polling is inefficient. Memorystore Redis is ephemeral and in-memory. Source: Check Source
Question #5 - Managing Application Performance Monitoring
A developer sees overall p95 latency spike on a microservices application but cannot identify which downstream call is the culprit.
Which Google Cloud service surfaces per-call spans across services?
A) Cloud Trace
B) Cloud Storage object versioning
C) Artifact Registry scanning
D) Cloud DNS
Show solution
Correct answers: A – Explanation:
Cloud Trace captures distributed spans across services, which is what is needed to find the slow hop. Object versioning is storage-focused. Artifact Registry scanning checks container vulnerabilities. Cloud DNS is resolution. Source: Check Source
Question #6 - Integrating Google Cloud Services
A Cloud Run service must react to Cloud Audit Log events across many Google services in near real time, without polling.
Which Google Cloud glue is most appropriate?
A) Cloud DNS policies
B) A static BigQuery view
C) Eventarc triggers
D) Cron in a Compute Engine VM
Show solution
Correct answers: C – Explanation:
Eventarc delivers events from sources like Cloud Audit Logs to Cloud Run, which is exactly this pattern. Cloud DNS, BigQuery views, and cron jobs do not deliver event streams to services. Source: Check Source
Question #7 - Deploying Applications
A team wants to roll out a new Cloud Run revision to 10% of traffic for 30 minutes before promoting to 100%.
Which Cloud Run feature most directly supports this?
A) A Cloud Storage bucket canary flag
B) Traffic splitting across revisions
C) Deleting the prior revision immediately
D) Manual DNS swaps only
Show solution
Correct answers: B – Explanation:
Cloud Run’s traffic splitting assigns percentages across revisions, enabling canary and blue-green deployments. A bucket flag does not control traffic. Deleting the prior revision is a hard cutover. Manual DNS is coarse and slow. Source: Check Source
Question #8 - Designing Highly Scalable, Available, and Reliable Cloud-Native Applications
A service sees 10x traffic on Black Friday and idles most of the year. The team wants the simplest autoscaling story with no fleet management.
Which runtime fits best?
A) A fixed-size Compute Engine MIG sized for Black Friday
B) A single Compute Engine VM with big RAM
C) GKE with manual node pool sizing
D) Cloud Run or Cloud Functions with concurrency-based autoscaling
Show solution
Correct answers: D – Explanation:
Serverless runtimes like Cloud Run and Cloud Functions autoscale on request volume with no fleet to manage. An always-on MIG sized for peak is wasteful. A single large VM does not scale horizontally. Manual GKE node sizing is the opposite of simple. Source: Check Source
Question #9 - Building and Testing Applications
A dev team wants a single managed place to store container images and language packages with vulnerability scanning.
Which Google Cloud service fits?
A) Artifact Registry
B) Cloud Source Repositories
C) Cloud Storage buckets as a registry
D) Bigtable for artifacts
Show solution
Correct answers: A – Explanation:
Artifact Registry stores container images and language packages with integrated vulnerability scanning. Cloud Source Repositories store source code. Cloud Storage is generic object storage. Bigtable is a wide-column store. Source: Check Source
Question #10 - Managing Application Performance Monitoring
A developer is debating whether to ship a risky feature now or wait. The SRE lead notes that the service has consumed 90% of its monthly error budget.
Which decision aligns with SRE practice?
A) Ship the risky feature and hope for no incidents
B) Delay the risky change; the team is near its monthly error budget
C) Disable monitoring so the budget resets
D) Remove the SLO to avoid constraint
Show solution
Correct answers: B – Explanation:
SRE practice says when the error budget is nearly exhausted, teams slow risky releases. Shipping anyway invites SLO breach. Disabling monitoring hides reality. Removing the SLO abandons the contract with users. Source: Check Source
Get 1246+ more questions with source-linked explanations
Every answer traces to the exact Google documentation page — so you learn from the source, not just memorize answers.
Exam mode & learn mode · Score by objective · Updated April 21, 2026
Learn more...
What the Cloud Developer exam measures
- Designing Highly Scalable, Available, and Reliable Cloud-Native Applications (25%): Apply Google Cloud practices to event-driven patterns, idempotency, autoscaling, graceful degradation.
- Building and Testing Applications (20%): Apply Google Cloud practices to Cloud Build, Artifact Registry, unit and integration testing, Cloud Code.
- Deploying Applications (20%): Apply Google Cloud practices to Cloud Run, Google Kubernetes Engine, App Engine, blue-green and canary patterns.
- Integrating Google Cloud Services (20%): Apply Google Cloud practices to Pub/Sub, Cloud Tasks, Eventarc, Cloud Endpoints, secrets management.
- Managing Application Performance Monitoring (15%): Apply Google Cloud practices to Cloud Trace, Cloud Profiler, Cloud Logging, error budgets.
How to prepare for this exam
- Review the Professional Cloud Developer official exam guide end to end before you commit a study plan, so every later hour is spent against the published blueprint.
- Complete the relevant Google Cloud Skills Boost learning path and treat its labs as non-optional rather than extra credit.
- Get hands-on practice in Qwiklabs sandbox, repeating the same tasks from memory until configuration feels routine.
- Apply what you learn in real-world project experience — your day job, a volunteer project, or an open-source contribution — so the concepts stick.
- Master one objective at a time, starting with the highest-weighted domain on the blueprint and moving down from there.
- Use PowerKram learn mode with feedback and sourced links to close gaps while the answer rationale is still fresh.
- Finish with PowerKram exam mode across all objectives under realistic time pressure before you book the real exam.
Career paths and salary outlook
Holding the Professional Cloud Developer certification typically supports roles such as:
- Cloud Application Developer: roughly $ 110,000 to $160,000 USD per year in the US market (range varies by region, years of experience, and specialization). See current data on Glassdoor.
- Backend Engineer (GCP): roughly $ 120,000 to $170,000 USD per year in the US market (range varies by region, years of experience, and specialization). See current data on Levels.fyi.
- Platform Engineer: roughly $ 130,000 to $180,000 USD per year in the US market (range varies by region, years of experience, and specialization). See current data on Payscale.
Official resources
Work directly from Google’s own preparation resources and treat third-party content as a supplement: