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.
A solutions architect is designing a benchmarking strategy to evaluate whether a newly proposed managed database cluster on Google Cloud can meet a strict transactional throughput and latency Service Level Objective (SLO). The benchmark requires running multi-threaded OLTP test suites (such as pgbench or HammerDB) against the database instance over private networking.
To ensure the benchmark produces valid baseline measurements and avoids client-side bottlenecks or networking anomalies, which setup should the architect implement?
This configuration deploys the load generation client in a different geographic region and points the benchmark workload across public internet gateways using external IP addresses.
The team were specifically conducting end-user global access testing or validating cross-region disaster recovery replication lag.
Infrastructure benchmarking focuses on measuring the raw capabilities and limits of Google Cloud resources under predictable, controlled network conditions.
This option runs the benchmark driver locally on an on-premises machine, sending all synthetic database transactions across a hybrid Cloud VPN tunnel into the Google Cloud VPC.
The primary requirement was to evaluate the bandwidth and packet-processing limits of the hybrid Cloud VPN gateway itself under simulated database replication traffic.
To test both infrastructure baselines and user experience, teams first benchmark resources internally within the cloud zone/VPC, and subsequently perform synthetic user journey testing using Cloud Monitoring synthetic monitors.
This option provisions a shared-core or minimal compute instance (e2-micro) to host and run the benchmarking suite against the database to reduce infrastructure expenses during test execution.
e2-micro instance only provides fractional vCPU bursts and minimal memory, causing the benchmark harness to become severely CPU-bound almost immediately.The goal were to perform a simple end-to-end connectivity check or run lightweight administrative validation scripts rather than rigorous load benchmarking.
Cost optimization during benchmarking should be achieved by deleting benchmark instances immediately after test completion rather than under-provisioning the client harness.
This architecture establishes a controlled, high-performance load generation environment for database and infrastructure benchmarking. By placing a sufficiently sized Compute Engine virtual machine in the exact same failure domain (zone), region, and Virtual Private Cloud (VPC) network as the database instance, the test isolates the database subsystem under test and eliminates confounding variables.
e2-standard-32 instance) ensures the client itself never saturates its CPU or network stack during load generation.Compared to executing tests over WAN, VPN, or across regions, running a dedicated, high-vCPU client within the same zone and VPC gives the cleanest signal for establishing baseline database performance and verifying architectural SLOs.
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