Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Developer practice bank has an additional 608 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A software engineering team is architecting a new analytics application on Google Cloud consisting of two components with different infrastructure requirements:
Which compute strategy should the team choose?
Compute Engine is Google Cloud's Infrastructure as a Service (IaaS) platform that provides self-managed virtual machine (VM) and bare metal instances. Cloud Run is a fully managed serverless platform designed to execute stateless containerized applications without requiring server or cluster administration.
Matching each workload to the appropriate level of the control-versus-abstraction spectrum avoids unnecessary operational burden. Using Compute Engine strictly for the kernel-dependent collector preserves full administrative flexibility, while deploying the stateless API on Cloud Run minimizes daily maintenance and cost compared to managing VMs or Kubernetes clusters.
GKE Autopilot is a managed Kubernetes mode of operation in which Google provisions, configures, and scales the underlying cluster nodes and security posture automatically, allowing teams to pay only for the CPU, memory, and storage requested by their Pods.
While Compute Engine satisfies the Telemetry Collector's kernel requirements, deploying a single stateless HTTP Webhook API on GKE Autopilot introduces unnecessary Kubernetes complexity and operational overhead. Unlike Cloud Run, GKE requires managing Kubernetes manifests, cluster lifecycles, and per-cluster hourly management fees, and it is less streamlined for instant scale-to-zero HTTP request handling.
The Webhook API were part of a complex microservices architecture, required stateful persistent volumes, or had an explicit dependency on Kubernetes APIs and Pod-level scheduling controls.
Unless a workload specifically needs Kubernetes features, serverless platforms like Cloud Run are recommended first because they require less self-managed configuration and deliver lower operational overhead for stateless HTTP services.
Google Kubernetes Engine (GKE) in Standard mode is a Container as a Service (CaaS) platform where Google manages the Kubernetes control plane, while you manage the cluster node pools, workload scheduling, and cluster configurations.
Although GKE Standard provides greater flexibility than GKE Autopilot, Google still manages the underlying host operating system on cluster nodes. It is not designed for workloads that require modifying boot files, loading arbitrary custom Linux kernel modules, or exercising full IaaS-level administrative control over a virtual server's OS.
The Telemetry Collector were a standard containerized service that needed custom Kubernetes node configurations, specialized DaemonSets, or complex multi-container orchestration rather than custom kernel settings and boot-file control.
A common misconception is that managing node pools in GKE Standard provides the same OS and kernel autonomy as Compute Engine. When a workload explicitly demands specialized kernel settings or custom OS modifications, an IaaS platform like Compute Engine is required.
Cloud Run is a fully managed serverless application platform that executes containerized web services, background jobs, and event-driven functions directly on Google's infrastructure without exposing underlying virtual machines or clusters.
Cloud Run is an ideal choice for the stateless Webhook API, but it cannot support the Telemetry Collector. In a serverless execution environment, Google completely manages the underlying OS, kernel, and runtime infrastructure. Developers have no access to customize the host kernel, load custom kernel modules, or modify boot files.
Both the Telemetry Collector and the Webhook API were standard stateless containerized workloads that relied solely on user-space libraries without requiring host kernel modifications or custom boot configurations.
Packaging an application inside a Docker container allows you to customize user-space dependencies, but containers still depend on the host kernel. When kernel-level customization is mandatory, you must move down the abstraction stack to Compute Engine.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated