GCP Professional Data Engineer Practice Question

Your organization is containerizing a portfolio of Java microservices that must run on both Google Cloud and an existing Amazon EKS environment for disaster-recovery purposes. Leadership insists that the same CI/CD pipeline and deployment artifacts be reused across clouds, and the operations team wants to avoid any Google-specific objects that could complicate failover. Which approach will best satisfy these portability requirements while using Google Kubernetes Engine (GKE) in production?

Migrate the applications to App Engine standard on GCP and to AWS Elastic Beanstalk for failover, accepting minor code rewrites for each environment.
Adopt GKE Autopilot with Google Cloud Load Balancing and Multi-Cluster Ingress; export GKE-generated configs and adjust them manually before applying to EKS.
Deploy the containers to Cloud Run on GCP and to AWS Fargate on demand, rewriting deployment descriptors for each platform's serverless model.
Use Cloud Build to produce OCI-compliant images, push them to Artifact Registry, and deploy the same Kubernetes Deployments, Services, and standard Ingress resources to both GKE and EKS, running a community NGINX Ingress controller on each cluster.

Report Issue

Answer Description

GKE exposes the same upstream Kubernetes APIs that Amazon EKS supports, so workloads defined with standard Kubernetes objects (Deployments, Services, Ingress) can be moved between the two with no code changes. Building OCI-compliant container images in Cloud Build and storing them in Artifact Registry makes the images accessible from any registry-compatible cluster, including EKS. Using a vendor-neutral Ingress controller such as community NGINX maintains portability, whereas Cloud Load Balancing features like Multi-Cluster Ingress, Cloud Run, or App Engine introduce GCP-specific dependencies. Re-implementing services in Cloud Functions, Cloud Run, or platform-specific PaaS offerings would break the "single artifact, multi-cloud" requirement and increase operational overhead.

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.