Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Professional Cloud Architect practice bank has 2,335 more like it. Try 10 free: instant score, no account.
An AI research team is deploying multi-node large language model (LLM) fine-tuning jobs that take between 3 and 5 days to complete. The workload has the following architectural requirements:
Which AI Hypercomputer consumption option should you recommend?
Spot VMs provide excess Compute Engine capacity at deeply discounted rates of up to 60% to 91% compared to standard on-demand pricing. In exchange for lower pricing, Google Cloud retains the right to preempt and reclaim instances at any time with a 30-second notice.
Spot VMs would be the optimal choice for fault-tolerant batch processing, continuous integration (CI/CD) pipelines, or training jobs with frequent multi-tier checkpointing where interrupted instances can resume seamlessly without causing total job failure.
Flex-start is an AI Hypercomputer and Google Cloud compute consumption model powered by the Dynamic Workload Scheduler (DWS). It allows teams to request dense compute capacity for short-duration workloads lasting up to seven days without committing to a long-term reservation.
Flex-start is specifically designed for non-fault-tolerant, short-duration training and fine-tuning jobs that have flexible start windows. It provides the low-latency network topology of reservations and significant cost savings without the ongoing financial overhead of dedicated calendar reservations.
Future reservations in calendar mode allow users to view future hardware availability across Google Cloud regions and reserve up to 80 GPU VMs for a duration of up to 90 days. Upon approval, Compute Engine guarantees capacity delivery on the specified start date.
The team had fixed training schedules, required strict delivery date assurance for upcoming deadlines, and intended to keep the GPU infrastructure continuously utilized throughout the reserved window to justify the reservation billing.
Standard on-demand provisioning allocates Compute Engine virtual machines immediately from available general pools on a standard pay-as-you-go (PAYG) basis, without reservations or dynamic scheduling queues.
The workload required immediate startup for ad-hoc interactive testing, used non-scarce CPU machine types, or could not accept queued dispatching.
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