Free Google Cloud Certified - Associate Cloud Engineer practice — 6 questions on Deploying and Implementing a Cloud Solution, with explanations. No sign-up.
Full 12-question mixed test →
Question 1 of 6 · Deploying and Implementing a Cloud Solution
A managed instance group autoscales based on CPU utilization. During load testing, engineers notice the autoscaler adds instances too aggressively immediately after a scale-up event, causing thrashing. Which setting should be adjusted to prevent the autoscaler from re-evaluating metrics too soon after a scaling action?
The MIG autoscaler's --cool-down-period (default 60 seconds) controls how long the autoscaler waits after adding an instance before it evaluates metrics again, preventing rapid repeated scale-up actions.
Question 2 of 6 · Deploying and Implementing a Cloud Solution
A GKE cluster runs a stateless web application across a node pool with 10 nodes. The platform team must upgrade the node pool to a new GKE version, keep the application fully available throughout, and minimize the number of extra nodes provisioned during the upgrade. Which upgrade strategy should they configure?
A surge upgrade with max-surge=1 and max-unavailable=0 replaces nodes one at a time, adding only a single extra node at any point while guaranteeing no reduction in available capacity, meeting both the zero-downtime and minimal-extra-nodes requirements.
Question 3 of 6 · Deploying and Implementing a Cloud Solution
A team needs to run a nightly batch process that processes a large CSV file, runs for up to 90 minutes, does not serve HTTP traffic, must exit with a status code, and should be triggered on a schedule via Cloud Scheduler. Which deployment approach is the best fit?
Cloud Run jobs are purpose-built for run-to-completion, non-HTTP workloads that can run up to 24 hours, exit with a status code, and are commonly triggered on a schedule by invoking them from Cloud Scheduler with an OIDC token.
Question 4 of 6 · Deploying and Implementing a Cloud Solution
A Cloud SQL for PostgreSQL read replica needs to be promoted to a standalone, writable primary instance because the original primary instance suffered an unrecoverable regional outage. Which command accomplishes this?
gcloud sql instances promote-replica converts an existing read replica into a standalone instance that can accept writes, breaking the replication relationship with the (now unavailable) primary.
Question 5 of 6 · Deploying and Implementing a Cloud Solution
A publisher application must guarantee that messages tagged with the same order ID are delivered to subscribers in the exact order they were published, even though multiple subscriber processes pull concurrently from the subscription. Which combination of settings should be configured when deploying the Pub/Sub topic and subscription?
Pub/Sub delivers ordered messages only when the publisher sets a consistent ordering key on related messages and the subscription has message ordering enabled; without both, order is not guaranteed even with a single publisher.
Question 6 of 6 · Deploying and Implementing a Cloud Solution
When comparing Google Cloud Deployment Manager to Terraform for provisioning Google Cloud resources, which statement accurately describes their state management?
Deployment Manager is a Google Cloud-native service that tracks the state of deployments internally via the Deployment Manager API, whereas Terraform is provider-agnostic and requires you to configure and manage a state file, commonly stored remotely in a GCS bucket backend for team use.
Ready for the real thing?
The full course: two full-length practice tests, video lessons for every exam domain, hands-on labs and detailed explanations.
$99.99$34.99 with code FREETEST33 — valid through September 14.