GCP Professional Cloud Architect Practice Question

Your company runs a Python Flask API in Docker containers on a small on-premises Kubernetes cluster. Traffic is highly bursty, often dropping to zero overnight, and the operations team struggles with cluster patching and scaling. Leadership wants to migrate this workload to Google Cloud within two months, reduce operational overhead, pay only for actual usage, and avoid managing infrastructure, while keeping container-based deployment to allow future CI/CD enhancements. Which migration approach best aligns with a cloud-first design philosophy?

Use Migrate to VMs to move the on-premises Kubernetes worker nodes into Compute Engine and continue operating the cluster yourself.
Rehost the containers unchanged on a managed instance group of Compute Engine VMs fronted by an external HTTP(S) load balancer.
Package the application image and deploy it to Cloud Run fully managed, relying on its automatic scaling and per-request billing.
Create a Google Kubernetes Engine (GKE) Standard cluster in one region and deploy the containers with Horizontal Pod Autoscaling enabled.

Report Issue

Answer Description

A cloud-first approach favors managed, serverless platforms that minimize infrastructure management and provide granular, consumption-based pricing. Deploying the existing container image to Cloud Run satisfies these goals because Cloud Run is fully managed, scales automatically down to zero during idle periods, and bills only for the exact resources consumed per request. In contrast, Compute Engine managed instance groups, a migrated self-managed Kubernetes cluster, or a new GKE Standard cluster still require the organization to patch, secure, and capacity-plan VM nodes, which contradicts the requirement to avoid infrastructure management and pay strictly per use.

Ask Bash

Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.