Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Database Engineer practice bank has an additional 769 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A retail company runs a production Cloud SQL Enterprise edition instance experiencing performance bottlenecks during peak business hours. Monitoring shows the following metrics and configuration:
The team needs to vertically scale the instance to maximize storage throughput and IOPS while avoiding business-hours downtime and conflicts with scheduled maintenance.
Which scaling and operational strategy should the team implement?
Cloud SQL offers HDD (hard disk drive) storage as a lower-cost alternative to SSD (solid-state drive) storage for specific workloads, while memory resizing adjusts the RAM allocation of the database VM.
The workload involved an archival or batch-processing database larger than 10 TB that was insensitive to latency, CPU utilization was low, and daytime downtime was acceptable.
Coordinated vertical scaling in Cloud SQL involves independently adjusting storage capacity and compute resources (vCPU and RAM) according to their distinct operational behaviors. Because Cloud SQL separates storage resizing from compute machine-type modifications, administrators can expand persistent disk capacity online while scheduling compute upgrades during controlled off-peak periods.
Unlike approaches that rely on automatic storage increments, downgrade to slower storage tiers, or collide with maintenance windows, this phased strategy resolves urgent I/O pressure immediately without downtime and executes the disruptive compute restart safely.
Automatic storage increases monitor available disk space every 30 seconds and permanently add a small increment of storage only when free space drops below a specific threshold. Meanwhile, vCPU scaling changes the underlying virtual machine shape allocated to the database instance.
The goal was strictly to prevent an out-of-disk outage from gradual organic data growth over time, and an immediate restart during business hours was acceptable to the application.
This approach leaves the storage capacity untouched while attempting to piggyback a manual vCPU upgrade directly onto Cloud SQL's automated weekly maintenance window.
Disk IOPS and free space were already well within healthy thresholds (under 80% used with sufficient throughput), and the vCPU resize was scheduled cleanly outside the maintenance window rather than concurrently with it.
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