Cloud Engineer Intermediate Practice Exam: Medium Difficulty 2025

Your organization has a Google Cloud Organization with multiple teams. You need a new environment for a project that must (1) separate billing from other teams, (2) enforce a policy that blocks external IP addresses on VMs, and (3) allow the networking team to manage shared VPC networks. What should you do?

You are onboarding a new application team. They must deploy resources using Infrastructure as Code, but they should not be able to create or modify IAM policies. They also need to create and manage Compute Engine instances and Cloud Storage buckets in a single project. What is the best approach?

A web application runs on managed instance groups (MIGs) in two zones within one region. The application must survive a single zone failure and support HTTP(S) traffic with a single anycast IP and health-based failover. Which architecture should you implement?

Your team needs to run batch jobs nightly. Jobs read from a Cloud Storage bucket, perform CPU-intensive processing for about 30 minutes, and write results back to Cloud Storage. You want minimal operational overhead and the ability to scale up to hundreds of concurrent jobs when needed. What should you use?

You are migrating an on-prem application to Compute Engine. The application requires a fixed internal IP address, and it must start automatically after host maintenance events. The VM must also be placed in a specific zone due to a dependency on zonal resources. What should you do?

A team is deploying a containerized API to Cloud Run. The service must privately access a Cloud SQL instance without exposing the database to the public internet. The team also wants to avoid managing VM-based proxy components. What should you do?

You need to process messages from a Pub/Sub topic and write results to BigQuery. Processing must be resilient to transient failures, and you need at-least-once delivery semantics. You want a managed solution with minimal custom retry logic. What should you choose?

Your application runs on a managed instance group behind an HTTP(S) Load Balancer. Users report intermittent 500 errors. You need to identify whether errors are coming from the load balancer, the backend instances, or the application, and you want to correlate requests across components. What should you do?

You manage a production project with strict uptime requirements. You need to be alerted when a zonal managed instance group’s healthy instances drop below a threshold for more than 5 minutes, and the alert should notify an on-call channel. What should you do?

A Cloud Storage bucket contains sensitive files. Only a specific service account used by a Dataflow job should be able to read objects, and users should not be able to make the bucket publicly accessible now or in the future. What should you implement?