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 organization is designing the cloud architecture for a new event-driven order processing application. During planning, the architecture and finance teams establish the following constraints:
Which Google Cloud compute option and configuration best addresses these constraints?
Sole-tenant nodes are dedicated physical Compute Engine servers reserved exclusively for a single organization's virtual machines, preventing hardware sharing with other cloud customers. CPU overcommit is a sole-tenant configuration feature (adjustable between 1.0 and 2.0) that allows VMs on the dedicated node to share physical CPU resources and burst above their baseline allocation.
This configuration would be ideal if the organization needed to migrate enterprise Windows Server workloads with existing per-core BYOL agreements or strict regulatory hardware isolation mandates, where the VMs are generally underutilized on the dedicated host but occasionally need to burst CPU usage.
While CPU overcommit improves resource density on dedicated physical hosts for bursty virtual machines, it operates within the fixed capacity of provisioned sole-tenant nodes and does not eliminate baseline hardware costs or OS management overhead.
Spot VMs are highly discounted, ephemeral Compute Engine virtual machine instances that utilize spare Google Cloud capacity, which Google Cloud can reclaim at any time if the capacity is needed elsewhere. Custom machine types allow administrators to tailor the exact number of vCPUs and amount of memory for a VM to match specific workload profiles and avoid overprovisioning.
This solution would be appropriate if the company needed to run non-critical, fault-tolerant batch processing jobs or experimental development workloads with unusual CPU-to-memory ratios, where low cost is the primary objective and sudden instance preemption is acceptable.
In a well-architected cloud environment, Spot VMs are paired with stateless, fault-tolerant processing pipelines or sandbox testing environments, whereas customer-facing, event-driven applications rely on managed serverless platforms like Cloud Run for reliable, instant scaling.
Cloud Run is a fully managed serverless compute platform that enables teams to run containerized applications without provisioning, configuring, or maintaining underlying virtual machine infrastructure. By combining Cloud Run with built-in autoscaling and maximum scaling limits, organizations can dynamically adjust compute capacity in real time based on incoming requests or events.

Compared to virtual machine-based solutions like Compute Engine, Cloud Run removes the operational overhead of OS management and prevents paying for idle, overprovisioned capacity. It provides the ideal balance between dynamic performance, strict cost governance, and minimal operational complexity for event-driven workloads.
Compute Engine is an Infrastructure as a Service (IaaS) offering that provides configurable virtual machines (VMs) with full control over the operating system and runtime environment. A 3-year resource-based committed use discount (CUD) is a pricing model that provides a deep discount on specific Compute Engine resources in a designated region in exchange for a three-year contract.
This option would be the right choice if the organization were migrating a predictable, steady-state production workload that runs continuously at stable utilization levels in a single region, and the engineering team required granular operational control over the underlying operating system to achieve the lowest long-term total cost of ownership (TCO).
While Compute Engine also offers flexible CUDs that apply across multiple machine types and regions based on hourly spend, any committed use discount is fundamentally designed for baseline, steady-state consumption rather than highly erratic, event-driven workloads.
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