GCP Cloud Digital Leader Practice Question

Your company runs a customer-facing web application on Compute Engine virtual machines located in a single zone. Leadership asks you to redesign the deployment so the service remains available even if that zone goes offline. Which Google Cloud design change best meets this requirement for high availability and fault tolerance?

Switch the virtual machines to preemptible instances within the same zone so they will restart automatically if interrupted.
Purchase committed use discounts for the existing virtual machines to ensure capacity in the current zone.
Migrate the application to a regional managed instance group and place it behind a global HTTP(S) Load Balancer.
Replace the current deployment with a single larger virtual machine that uses local SSDs in the existing zone.

Report Issue

Answer Description

Deploying the application in a regional managed instance group (MIG) spreads virtual machine instances across multiple zones within the same region. When the MIG is placed behind a global external HTTP(S) Load Balancer, user traffic is automatically routed only to healthy instances. If one zone becomes unavailable, the load balancer directs requests to instances in the remaining zones, allowing the service to continue operating without manual intervention. Committed use discounts reduce cost but do not add resilience. Preemptible VMs are inexpensive but can be terminated at any time and are not suited to improving availability. Consolidating the workload on a single larger VM in the same zone actually increases the blast radius of a failure rather than reducing it.

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.