Google Professional Cloud Architect Practice Test 53
Question 520
At Medical translation nonprofit, the speech-to-text triage workload has a documented risk: clinics need transcripts of spoken patient questions and multilingual summaries. The review found that audio samples vary in background noise and language code switching. The architecture must also ensure that patient data must be handled under consent and short retention. The team must deliver the approved control safely. Which deployment sequence is most appropriate?
- Roll out authorized audio processing with regional retention controls
- Use an appropriate speech recognition API with separate governed translation and summarization stages
- Use an image-labeling API as the only audio transcriber
- Monitor failed transcriptions, language bias and data-deletion completions
Explanation
Correct Answer
A:
Roll out authorized audio processing with regional retention controls is a controlled means to introduce the approved architecture while meeting the requirement that patient data must be handled under consent and short retention. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
B:
The alternative “Use an appropriate speech recognition API with separate governed translation and summarization stages” describes the chosen design, but not how it is introduced safely. In the reported observation, audio samples vary in background noise and language code switching. The option “Roll out authorized audio processing with regional retention controls” instead addresses the requested controlled implementation.
C:
An image classification service does not provide speech transcripts. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that patient data must be handled under consent and short retention.
D:
The alternative “Monitor failed transcriptions, language bias and data-deletion completions” addresses later monitoring rather than the actual controlled deployment. In the reported observation, audio samples vary in background noise and language code switching. The option “Roll out authorized audio processing with regional retention controls” instead addresses the requested controlled implementation.
Question 521
At Financial remittance company, the card token ledger workload has a documented risk: a transactional SQL database runs out of horizontal write capacity across continents. The review found that design review finds one region primary is a regulatory single point of failure. The architecture must also ensure that ledger updates require strongly consistent multi-region transactions. What follow-up metric or alert will identify emerging failures after the initial acceptance test?
- Create independent writable ledgers and reconcile disputed balances weekly
- Design multi-region Spanner placement with bounded consistency and local compliance review
- Reconcile unique transfer IDs and fail a region during concurrent writes
- Track transaction aborts, commit latency and regional availability
Explanation
Correct Answer
D:
Track transaction aborts, commit latency and regional availability reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (a transactional SQL database runs out of horizontal write capacity across continents) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
Deferred conflict reconciliation is unacceptable for money movement. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that ledger updates require strongly consistent multi-region transactions.
B:
The alternative “Design multi-region Spanner placement with bounded consistency and local compliance review” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, design review finds one region primary is a regulatory single point of failure. The option “Track transaction aborts, commit latency and regional availability” instead addresses the requested ongoing operational assurance.
C:
The alternative “Reconcile unique transfer IDs and fail a region during concurrent writes” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, design review finds one region primary is a regulatory single point of failure. The option “Track transaction aborts, commit latency and regional availability” instead addresses the requested ongoing operational assurance.
Question 522
Consider National investment office and its emergency break-glass access workload. The issue is that a single admin group controls production recovery and can also edit security logs; the current observation is that tabletop shows no independent emergency account if primary IdP fails. The team must ensure that recovery access must be auditable and tightly time-bound. Which acceptance test should demonstrate both protection against the failure and continued authorized service?
- Share one break-glass password in the team chat channel
- Provision sealed emergency accounts with quarterly practice and credential rotation
- Map critical emergency actions, alternative identities and audit paths
- Simulate IdP outage and confirm authorized break-glass restores service with logs
Explanation
Correct Answer
D:
Simulate IdP outage and confirm authorized break-glass restores service with logs is a direct acceptance test of the path implicated by tabletop shows no independent emergency account if primary IdP fails. It checks that the original risk (a single admin group controls production recovery and can also edit security logs) has actually been controlled and tests that permitted operations still meet the requirement that recovery access must be auditable and tightly time-bound.
Incorrect Answers
A:
A widely shared emergency secret is not protected or attributable. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that recovery access must be auditable and tightly time-bound.
B:
The alternative “Provision sealed emergency accounts with quarterly practice and credential rotation” documents change execution but not its operational acceptance result. In the reported observation, tabletop shows no independent emergency account if primary IdP fails. The option “Simulate IdP outage and confirm authorized break-glass restores service with logs” instead addresses the requested acceptance verification.
C:
The alternative “Map critical emergency actions, alternative identities and audit paths” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, tabletop shows no independent emergency account if primary IdP fails. The option “Simulate IdP outage and confirm authorized break-glass restores service with logs” instead addresses the requested acceptance verification.
Question 523
Case study: Subscription video startup: BigQuery warehouse scan cost
Subscription video startup relies on BigQuery warehouse scan cost to deliver its operational commitments. The current architectural problem is that queries scan every historical event for simple daily subscriber counts. The team has recorded this evidence: query plan shows 5TB scanned despite filtering for yesterday. Any proposed response must preserve the requirement that marketing results must arrive within its daily budget.
At Subscription video startup, the Google Cloud design for BigQuery warehouse scan cost is affected because queries scan every historical event for simple daily subscriber counts. A review found that query plan shows 5TB scanned despite filtering for yesterday. The proposed response must account for the requirement that marketing results must arrive within its daily budget. Which controlled implementation should be used to preserve operational continuity during the change?
- Replace predicates with full-table SELECT star on each dashboard load
- Alert on nonpartitioned scans, query budgets and freshness
- Rewrite incremental pipelines and test a representative partitioned query
- Use date-partitioned tables and suitable clustering with governed query patterns
Explanation
Correct Answer
C:
Rewrite incremental pipelines and test a representative partitioned query is a controlled means to introduce the approved architecture while meeting the requirement that marketing results must arrive within its daily budget. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
Full scans multiply spend without improving daily freshness. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that marketing results must arrive within its daily budget.
B:
The alternative “Alert on nonpartitioned scans, query budgets and freshness” addresses later monitoring rather than the actual controlled deployment. In the reported observation, query plan shows 5TB scanned despite filtering for yesterday. The option “Rewrite incremental pipelines and test a representative partitioned query” instead addresses the requested controlled implementation.
D:
The alternative “Use date-partitioned tables and suitable clustering with governed query patterns” describes the chosen design, but not how it is introduced safely. In the reported observation, query plan shows 5TB scanned despite filtering for yesterday. The option “Rewrite incremental pipelines and test a representative partitioned query” instead addresses the requested controlled implementation.
Question 524
At Education research foundation, the Google Cloud design for multi-project packet filtering is affected because research projects inherit inconsistent firewall rules and centralized policy has no owner. A review found that network review finds SSH open to the internet from three experiment subnets. The proposed response must account for the requirement that a limited set of approved remote researchers still needs audited access. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Let every project administrator create a public any-any ingress rule
- Apply deny policies in monitored phases with approved access exceptions
- Inventory rule inheritance, exposure, identity and exception ownership
- Set hierarchy-level firewall controls with least-privilege project overrides
- Monitor firewall-rule drift and exception expiry by project
Explanation
Correct Answers
B:
Apply deny policies in monitored phases with approved access exceptions is a controlled means to introduce the approved architecture while meeting the requirement that a limited set of approved remote researchers still needs audited access. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption. This supplies the controlled-introduction half of the two-action package.
D:
Set hierarchy-level firewall controls with least-privilege project overrides targets the underlying issue (research projects inherit inconsistent firewall rules and centralized policy has no owner) rather than merely reporting symptoms. It also preserves the required outcome (a limited set of approved remote researchers still needs audited access), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
Incorrect Answers
A:
Unrestricted local firewalls defeat centralized protection. The shortcut does not supply the requested design and safe rollout for multi-project packet filtering, where network review finds SSH open to the internet from three experiment subnets.
C:
This diagnoses the original issue (research projects inherit inconsistent firewall rules and centralized policy has no owner), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
E:
Monitoring multi-project packet filtering for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: research projects inherit inconsistent firewall rules and centralized policy has no owner.
Question 525
Automotive safety modeling startup plans to update reinforcement model evaluation. Investigators found that evaluation report covers average accuracy but no rare safety-critical maneuvers, consistent with the reported problem that engineering wants to deploy a new model before proving corner-case behavior. The response must satisfy the requirement that safety review requires rigorous offline evaluation and rollback. The design review is over and production introduction is next. Which action is the best fit?
- Design model lifecycle gates with offline evaluation and isolated acceptance
- Approve the model solely because aggregate accuracy rose one percent
- Track model decision drift and post-release safety exceptions
- Introduce shadow inference before any controlled production promotion
Explanation
Correct Answer
D:
Introduce shadow inference before any controlled production promotion is a controlled means to introduce the approved architecture while meeting the requirement that safety review requires rigorous offline evaluation and rollback. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Design model lifecycle gates with offline evaluation and isolated acceptance” describes the chosen design, but not how it is introduced safely. In the reported observation, evaluation report covers average accuracy but no rare safety-critical maneuvers. The option “Introduce shadow inference before any controlled production promotion” instead addresses the requested controlled implementation.
B:
Mean accuracy can hide unacceptable safety-regression tails. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that safety review requires rigorous offline evaluation and rollback.
C:
The alternative “Track model decision drift and post-release safety exceptions” addresses later monitoring rather than the actual controlled deployment. In the reported observation, evaluation report covers average accuracy but no rare safety-critical maneuvers. The option “Introduce shadow inference before any controlled production promotion” instead addresses the requested controlled implementation.
Question 526
At National retail cooperative, the audit log preservation workload has a documented risk: a project owner can disable or delete local logging after an unauthorized configuration change. The review found that audit exercise shows log sinks owned by the affected project. The architecture must also ensure that investigations require independently retained administrative events. What follow-up metric or alert will identify emerging failures after the initial acceptance test?
- Store the only audit copy in a bucket controlled by the investigated project owner
- Aggregate audit logs into a protected central project with minimal deletion rights
- Watch logging exclusions, sink failures and retention violations
- Simulate project-level tampering and confirm central audit events survive
Explanation
Correct Answer
C:
Watch logging exclusions, sink failures and retention violations reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (a project owner can disable or delete local logging after an unauthorized configuration change) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
Shared deletion authority undermines evidentiary independence. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that investigations require independently retained administrative events.
B:
The alternative “Aggregate audit logs into a protected central project with minimal deletion rights” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, audit exercise shows log sinks owned by the affected project. The option “Watch logging exclusions, sink failures and retention violations” instead addresses the requested ongoing operational assurance.
D:
The alternative “Simulate project-level tampering and confirm central audit events survive” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, audit exercise shows log sinks owned by the affected project. The option “Watch logging exclusions, sink failures and retention violations” instead addresses the requested ongoing operational assurance.
Question 527
At Regional vehicle-parts supplier, the hybrid stock synchronization workload has a documented risk: headquarters reaches Google Cloud only through one IPsec tunnel that drops during maintenance. The review found that network log shows packets re-route onto a congested internet path after the gateway reboot. The architecture must also ensure that inventory updates need redundant private routing to two regions. What follow-up metric or alert will identify emerging failures after the initial acceptance test?
- Use a single static VPN gateway and advertise a default public route for all stock traffic
- Fail one peer and prove private inventory replication stays available
- Alert on tunnel status, route withdrawals and peer latency
- Use redundant HA VPN gateways with Cloud Router and independent peer paths
Explanation
Correct Answer
C:
Alert on tunnel status, route withdrawals and peer latency reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (headquarters reaches Google Cloud only through one IPsec tunnel that drops during maintenance) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
One tunnel and unscoped route reproduce the outage and security risk. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that inventory updates need redundant private routing to two regions.
B:
The alternative “Fail one peer and prove private inventory replication stays available” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, network log shows packets re-route onto a congested internet path after the gateway reboot. The option “Alert on tunnel status, route withdrawals and peer latency” instead addresses the requested ongoing operational assurance.
D:
The alternative “Use redundant HA VPN gateways with Cloud Router and independent peer paths” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, network log shows packets re-route onto a congested internet path after the gateway reboot. The option “Alert on tunnel status, route withdrawals and peer latency” instead addresses the requested ongoing operational assurance.
Question 528
Consider AI satellite imagery group and its short-lived GPU inference workload. The issue is that the product team leaves GPU instances running overnight for occasional image classification; the current observation is that billing shows high GPU idle cost with short bursts of queued requests. The team must ensure that inference can tolerate several seconds of cold-start delay. The change committee needs to understand the current system rather than install a fix yet. What should it do?
- Choose a scale-aware inference platform with workload-appropriate GPUs and queueing
- Measure model footprint, arrival patterns and accelerator warm-up cost
- Canary burst routing with model artifact caching and capped quotas
- Reserve the largest GPU fleet permanently for infrequent requests
Explanation
Correct Answer
B:
Measure model footprint, arrival patterns and accelerator warm-up cost is the appropriate diagnostic because the observation—billing shows high GPU idle cost with short bursts of queued requests—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (the product team leaves GPU instances running overnight for occasional image classification) before the design is approved.
Incorrect Answers
A:
The alternative “Choose a scale-aware inference platform with workload-appropriate GPUs and queueing” is target architecture selection; it does not establish why the observed path failed. In the reported observation, billing shows high GPU idle cost with short bursts of queued requests. The option “Measure model footprint, arrival patterns and accelerator warm-up cost” instead addresses the requested root-cause evidence.
C:
The alternative “Canary burst routing with model artifact caching and capped quotas” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, billing shows high GPU idle cost with short bursts of queued requests. The option “Measure model footprint, arrival patterns and accelerator warm-up cost” instead addresses the requested root-cause evidence.
D:
Always-on capacity wastes cost when demand is highly intermittent. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that inference can tolerate several seconds of cold-start delay.
Question 529
The external build pipeline identity environment at Online wholesale exchange faces this issue: a third-party CI runner receives a permanent Google service account key. Supporting evidence shows that audit shows key is copied across vendor workspaces. The architect must preserve the requirement that pipeline must deploy only into one dedicated release project. Before approving a solution, which analysis most directly clarifies the underlying dependency or control gap?
- Introduce federated deploy identities and revoke the permanent key
- Give the vendor a long-lived project Owner key to reduce setup work
- Inventory vendor identity claims, allowed repositories and target permissions
- Use OIDC-based Workload Identity Federation with repository-bound conditions
Explanation
Correct Answer
C:
Inventory vendor identity claims, allowed repositories and target permissions is the appropriate diagnostic because the observation—audit shows key is copied across vendor workspaces—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (a third-party CI runner receives a permanent Google service account key) before the design is approved.
Incorrect Answers
A:
The alternative “Introduce federated deploy identities and revoke the permanent key” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, audit shows key is copied across vendor workspaces. The option “Inventory vendor identity claims, allowed repositories and target permissions” instead addresses the requested root-cause evidence.
B:
Permanent broad keys magnify vendor compromise impact. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that pipeline must deploy only into one dedicated release project.
D:
The alternative “Use OIDC-based Workload Identity Federation with repository-bound conditions” is target architecture selection; it does not establish why the observed path failed. In the reported observation, audit shows key is copied across vendor workspaces. The option “Inventory vendor identity claims, allowed repositories and target permissions” instead addresses the requested root-cause evidence.
Question 530
Consider Retail airline reservations and its load-test realism workload. The issue is that capacity tests generate simple GET requests but real transactions require multiple write steps; the current observation is that peak production incident shows locks on seat inventory despite low GET latency. The team must ensure that holiday purchase flow must sustain transactional volume without overselling. Which continuous monitoring approach best reveals recurrence or changing dependencies?
- Declare peak readiness because static homepage GET requests are fast
- Design capacity and admission control around realistic end-to-end flows
- Run high-concurrency seat purchases and verify one valid reservation per seat
- Watch transaction lock waits and booking abandonment
Explanation
Correct Answer
D:
Watch transaction lock waits and booking abandonment reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (capacity tests generate simple GET requests but real transactions require multiple write steps) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
Read-only performance does not establish transactional capacity. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that holiday purchase flow must sustain transactional volume without overselling.
B:
The alternative “Design capacity and admission control around realistic end-to-end flows” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, peak production incident shows locks on seat inventory despite low GET latency. The option “Watch transaction lock waits and booking abandonment” instead addresses the requested ongoing operational assurance.
C:
The alternative “Run high-concurrency seat purchases and verify one valid reservation per seat” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, peak production incident shows locks on seat inventory despite low GET latency. The option “Watch transaction lock waits and booking abandonment” instead addresses the requested ongoing operational assurance.