Skip to content

GCP-PCD Google Cloud Professional Cloud Developer Practice Questions

Prepare for GCP-PCD with more than an answer.

156 questions in the full set12 sample questionsUpdated Mar 12, 2026

Unlock the full exam and previous versions

  • v1Google Cloud Professional Cloud Developer 156 questions Current
  • PCDLegacy Professional Cloud Developer 289 questions Locked
Exam fee
$200 USD
Level
Professional
Valid for
2 years
Domains covered on the exam 4
  1. Designing Highly Scalable, Available, and Reliable Cloud-Native Applications36%
  2. Building and Testing Applications23%
  3. Deploying Applications20%
  4. Integrating Applications with Google Cloud Services21%
  1. 1

    Your GKE cluster uses the Horizontal Pod Autoscaler (HPA) to scale a deployment based on CPU utilization. You notice that the pods scale up quickly during traffic spikes but take too long to scale down after the traffic subsides, causing unnecessary costs. Which HPA configuration parameter should you adjust?

    Show answer details

    Correct answer: B

    Kubernetes HPA (v2beta2 and later) supports a behavior field that allows fine-grained control over scaling policies. To address slow scale-down, you can adjust the stabilizationWindowSeconds or the scale-down policies (pods/percent per period) to allow faster termination of unnecessary pods.

  2. 2

    A developer investigates a performance issue where a specific API endpoint is slow. They want to visualize the latency of the request as it travels through multiple microservices (Frontend -> API Gateway -> Backend Service -> Database). Which Google Cloud service should they use to view this timeline?

    Show answer details

    Correct answer: B

    Cloud Trace is a distributed tracing system that collects latency data from your applications and displays it in the Google Cloud Console. It allows you to see the propagation of a request across microservices (waterfall view) and identify which step is causing the bottleneck.

  3. 3

    An application publishes messages to a Pub/Sub topic. The subscribers occasionally fail to process messages due to transient errors. To ensure that failed messages are not lost and can be analyzed later without blocking the subscription, what feature should you configure?

    Show answer details

    Correct answer: B

    A Dead Letter Topic allows Pub/Sub to forward messages that have exceeded the maximum delivery attempts to a separate topic. This prevents poison pill messages from blocking the queue while preserving them for debugging or manual reprocessing.

  4. 4

    A global gaming company is designing the backend for a new real-time multiplayer mobile game. The player profile service requires a database that offers strong consistency, horizontal scalability to handle millions of concurrent users globally, and 99.999% availability. The schema is relational, and players must see their stats updated immediately after a match, regardless of their region. Which Google Cloud storage solution should the architect select?

    Show answer details

    Correct answer: C

    Cloud Spanner is the only Google Cloud service that provides a relational database structure (SQL) combined with horizontal scalability and global strong consistency (using TrueTime). It is designed specifically for this use case (global scale, relational data, high availability up to 99.999%). Cloud SQL is regional (or requires manual sharding for writes). Bigtable is NoSQL, not relational.

  5. 5

    A developer is building a cloud-native application that processes uploaded images. The application needs to be event-driven. When a user uploads an image to a Cloud Storage bucket, a service should automatically be triggered to resize the image. The solution must be serverless and require minimal operational overhead. Which architecture should the developer implement?

    Show answer details

    Correct answer: B

    Eventarc is the preferred Google Cloud service for unifying event ingestion. It can listen for Cloud Storage events (via Audit Logs or direct notification) and push them to a Cloud Run service. This is a fully serverless, push-based architecture that avoids polling and operational overhead.

  6. 6

    You are migrating a legacy monolithic application to Google Cloud. The application requires a file system interface to read and write configuration files shared across multiple instances. You plan to containerize the application and deploy it to Google Kubernetes Engine (GKE). Which storage option should you mount to the GKE pods to support read/write many (RWX) access?

    Show answer details

    Correct answer: B

    Filestore provides fully managed NFS file servers on Google Cloud. GKE supports mounting Filestore instances as PersistentVolumes with ReadWriteMany (RWX) access mode, allowing multiple pods to read and write to the same shared file system simultaneously. Standard Persistent Disks generally support ReadWriteOnce (RWO).

Create an account to continue.