Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Cloud Digital Leader practice bank has an additional 3,070 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A retail company is migrating its e-commerce platform to Compute Engine virtual machines (VMs). The leadership team needs an infrastructure solution that addresses two operational challenges:
How does combining Compute Engine custom machine types with Managed Instance Group (MIG) autoscaling provide business value for this workload?
Placing capacity requests into a queue to wait for scarce resources describes time flexibility powered by the Dynamic Workload Scheduler (DWS) and the flex-start provisioning model in Compute Engine. Instead of requiring immediate, on-demand VM creation, this approach queues requests until high-demand hardware—such as GPUs or TPUs—becomes available.
The scenario involved running AI/ML model training, fine-tuning, or large-scale batch processing jobs that require specialized hardware accelerators and have flexible start times rather than real-time customer traffic demands.
While time flexibility is an important dimension of infrastructure optimization in Google Cloud, interactive web applications rely on real-time horizontal autoscaling and right-sized machine types to maintain immediate availability during load spikes.
Compute Engine provides machine type recommendations (also known as rightsizing recommendations) to help optimize the vCPU and memory allocation of virtual machines based on historical usage data over the previous 8 days. Applying a machine type recommendation resizes the capacity of individual VM instances (vertical scaling) without changing the total number of instances in a group.
The company was running a legacy workload tied to strict per-instance software licenses that prohibited adding more VM instances, and wanted to periodically right-size its existing VMs based on weekly utilization trends.
Understanding the distinction between vertical scaling (changing a VM's size via custom machine types or rightsizing recommendations) and horizontal autoscaling (changing the number of VMs in a MIG) is essential. Custom machine types establish the right baseline ratio for each VM, while MIG autoscaling varies the instance count in real time.
Instantly scaling compute resources without configuring fine-tuned scaling rules and automatically scaling down to zero instances during idle periods is the defining characteristic of serverless computing services, such as Cloud Run and Cloud Run functions. In a serverless model, the cloud provider abstracts away the underlying virtual machine infrastructure entirely.
CPU utilization, load balancing capacity, or Cloud Monitoring metrics) and minimum/maximum instance limits.The company was modernizing its application into stateless containers deployed on Cloud Run to achieve zero-configuration elasticity and zero cost during periods with no traffic.
Students often confuse the capabilities of IaaS autoscaling (Compute Engine MIGs) with serverless autoscaling (Cloud Run). While Compute Engine reduces cost during quiet periods by reducing the number of VMs, it requires explicit autoscaling policies and operates on VM instances rather than instant scale-to-zero execution environments.
Compute Engine custom machine types allow organizations to define the exact number of virtual CPUs (vCPUs) and the precise amount of memory required for an individual virtual machine (VM) instance. Managed Instance Group (MIG) autoscaling is an elastic infrastructure capability that dynamically adjusts the number of running VM instances in a group based on real-time load metrics, such as CPU utilization or load balancing serving capacity.
Combining vertical right-sizing (through custom machine types) with horizontal elasticity (through MIG autoscaling) maximizes price-performance in an Infrastructure as a Service (IaaS) environment. The organization pays only for the exact hardware resources each instance needs and runs only the exact number of instances required by current customer demand.

Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated