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 financial services company is deploying a new AlloyDB for PostgreSQL primary-only instance to support a hybrid transactional and analytical processing (HTAP) workload. Workload analysis reveals the following requirements:
Which compute sizing and utilization target should you select to maintain consistent throughput and avoid resource contention?
A 16 vCPU, 128 GB memory AlloyDB for PostgreSQL primary instance is sized for medium-to-large transactional and analytical workloads, supporting up to 8,000 transactions per second (TPS) (with a 65% read / 35% write ratio and 30% cached data) and an active working dataset of up to 1 TB. It also meets the minimum recommended compute threshold of 16 vCPUs or larger for primary-only HTAP deployments running the columnar engine.
max_connections limits.Smaller configurations (4 or 8 vCPUs) lack the TPS capacity, buffer pool memory, and columnar storage footprint required for a 700 GB HTAP workload, while much larger configurations (64 vCPUs) introduce unnecessary compute costs without proportional performance gains.
An 8 vCPU, 64 GB memory AlloyDB for PostgreSQL instance is designed for smaller OLTP workloads, supporting up to 4,000 TPS and an active working dataset of up to 500 GB. While it reduces upfront compute costs compared to larger tiers, it lacks the compute and memory capacity required for heavy HTAP operations.
The workload were a pure OLTP application generating fewer than 4,000 TPS with an active working dataset under 500 GB, and without the memory overhead of the columnar engine.
A 4 vCPU, 32 GB memory instance is sized for light transactional workloads up to 2,500 TPS and 250 GB of active working data. Raising the max_connections database flag allows more client sessions to open TCP connections simultaneously, but it does not add CPU cores or physical memory to execute those queries.
max_connections values are tied to machine size because smaller instances cannot sustain high concurrency.The workload generated under 2,500 TPS on a working dataset below 250 GB, and connection failures were caused strictly by idle sessions hitting a low default connection limit.
A 64 vCPU, 512 GB memory instance is an enterprise-scale compute shape built for massive workloads reaching up to 20,000 TPS and 8 TB of working data. Disabling the columnar engine keeps 100% of allocated memory available for the standard PostgreSQL buffer pool and session processes.
The database needed to sustain 15,000-20,000 TPS across an active working dataset of 5-8 TB for a purely transactional workload with minimal analytical aggregations.
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