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 software company is developing a new customer-facing application packaged in stateless containers. Currently, the engineering team hosts workloads on Compute Engine virtual machines (VMs), which requires them to spend significant time on administrative tasks. Leadership wants to modernize their compute approach to achieve the following goals:
Which compute solution should the company adopt?
Predictive autoscaling is an advanced scaling feature for Compute Engine managed instance groups (MIGs). It analyzes historical trends in metrics—such as CPU utilization—alongside real-time monitoring data to forecast future load and provision virtual machine instances ahead of expected traffic spikes.
Although predictive autoscaling improves application readiness for fluctuating traffic on VMs, it does not shift the workload away from infrastructure management:
This option would be appropriate if the company was required to keep its application on Compute Engine VMs due to strict kernel-level dependencies and needed to prevent latency during recurring, predictable traffic surges.
While predictive autoscaling enhances VM elasticity, serverless architectures like Cloud Run provide instant, built-in elasticity for stateless containers without requiring administrators to configure metric thresholds or forecast policies.
Cloud Run is a fully managed, serverless compute platform on Google Cloud designed to host and run stateless containers. It abstracts away all underlying infrastructure management, allowing organizations to deploy containerized applications directly without provisioning, configuring, or maintaining servers.
When evaluating compute abstractions, prioritizing managed services and serverless products maximizes business value. While Compute Engine VMs offer granular control, the operational overhead of maintaining and scaling those VMs increases TCO and slows down software delivery. Cloud Run provides the ideal abstraction for stateless containers by combining container portability with serverless operational efficiency.
A custom image is a preconfigured boot disk that contains application-specific software, dependencies, and system settings. An instance template combines that boot disk image with a specific machine type and other instance properties so administrators can rapidly provision identical Compute Engine virtual machines (VMs) or managed instance groups (MIGs).
While using custom images and templates speeds up the initial boot and provisioning time of VMs, it remains an Infrastructure as a Service (IaaS) approach that fails to remove administrative toil:
This option would be the right choice if the organization needed to run legacy, non-containerized applications on Compute Engine VMs and wanted to reduce instance startup times compared to installing dependencies dynamically via startup scripts.
A common misconception is that automating VM image creation eliminates management costs. In reality, maintaining an image-baking pipeline introduces recurring operational overhead that increases the total cost of ownership (TCO) compared to serverless platforms.
Running containers on Compute Engine involves provisioning standard virtual machine (VM) instances using a container-optimized operating system and deploying container images directly onto those VMs. This approach combines packaged container artifacts with traditional Infrastructure as a Service (IaaS) virtual machines.
Deploying containers directly onto individual VMs retains all the administrative burdens of managing virtual infrastructure:
This option would be the right choice if the organization had specialized compliance or technical requirements that demanded complete control over the underlying VM host configuration, custom kernel modules, or specific virtual machine hardware.
Packaging an application into a container makes it portable, but the operational agility gained depends heavily on the hosting platform. Hosting containers on unmanaged VMs forfeits the administrative benefits offered by Cloud Run or Google Kubernetes Engine (GKE) Autopilot.
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