Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Professional Cloud DevOps Engineer practice bank has 1,455 more like it. Try 10 free: instant score, no account.
Your team is architecting a compute environment for periodic AI/ML model fine-tuning and batch inference workloads. An analysis of the workload profile shows the following characteristics:
Which compute provisioning model and service configuration should you select to optimize costs while satisfying workload constraints?
Committed Use Discounts (CUDs) provide discounted hourly rates on Compute Engine resources in exchange for committing to pay for a minimum level of resource usage (vCPUs, RAM, or specific accelerators) continuously over a 1-year or 3-year term.
The workload represented a baseline, 24/7 production service with stable, predictable resource consumption over a multi-year horizon.
Spot VMs are surplus Compute Engine instances offered at the deepest available discount (ranging from 60% to 91% off standard on-demand rates). However, they come with the trade-off that Google Cloud can preemptively reclaim the compute capacity at any time with a 30-second notice.
Spot VMs would be the preferred choice if the workload was fully stateless or had lightweight checkpointing capabilities that allowed fast, low-overhead resumption after preemption.
Future Reservations in calendar mode allow organizations to reserve specific GPU or TPU machine capacity for a future date and time window of up to 90 days. Once approved, the capacity is guaranteed and delivered at the chosen start date.
The workload required guaranteed capacity on a fixed calendar date for sustained, continuous training operations spanning several weeks to 90 days.
Flex-start is a Google Cloud compute consumption and provisioning model designed for short-duration workloads (lasting up to seven days) that require compute resources and accelerators like GPUs. Under this model, Compute Engine schedules and provisions the requested compute instances as soon as capacity becomes available in the selected zone, providing access without requiring manual reservation negotiations.
Flex-start directly matches the combination of flexible start times, multi-day non-preemptible execution, and pay-as-you-go cost reduction, outperforming Spot VMs (which risk preemption) and fixed calendar reservations (which charge for idle reservation capacity).
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 Validated