Google Professional Cloud Architect Practice Test 32
Question 310
The customer loyalty relaunch environment at International fashion retailer faces this issue: a vendor proposal minimizes compute cost while ignoring peak checkout conversion. Supporting evidence shows that transaction dashboards show flash-sale load rises twelvefold for fifteen minutes. The architect must preserve the requirement that checkout must keep the payment processor independent of loyalty analytics. Before approving a solution, which analysis most directly clarifies the underlying dependency or control gap?
- Model conversion loss, flash-sale load profiles and payment dependencies
- Stage canary traffic with queue buffering and a payment-path rollback switch
- Synchronous-call the loyalty model before charging every order
- Separate checkout from asynchronous loyalty scoring with elastic stateless APIs
Explanation
Correct Answer
A:
Model conversion loss, flash-sale load profiles and payment dependencies is the appropriate diagnostic because the observation—transaction dashboards show flash-sale load rises twelvefold for fifteen minutes—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (a vendor proposal minimizes compute cost while ignoring peak checkout conversion) before the design is approved.
Incorrect Answers
B:
The alternative “Stage canary traffic with queue buffering and a payment-path rollback switch” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, transaction dashboards show flash-sale load rises twelvefold for fifteen minutes. The option “Model conversion loss, flash-sale load profiles and payment dependencies” instead addresses the requested root-cause evidence.
C:
Making payment dependent on optional scoring magnifies latency and outages. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that checkout must keep the payment processor independent of loyalty analytics.
D:
The alternative “Separate checkout from asynchronous loyalty scoring with elastic stateless APIs” is target architecture selection; it does not establish why the observed path failed. In the reported observation, transaction dashboards show flash-sale load rises twelvefold for fifteen minutes. The option “Model conversion loss, flash-sale load profiles and payment dependencies” instead addresses the requested root-cause evidence.
Question 311
At International telemedicine API, the client-library retry behavior workload has a documented risk: SDK integration retries non-idempotent appointment creation and doubles bookings. The review found that trace shows repeated POST requests after an upstream timeout. The architecture must also ensure that appointments must not be created twice for one patient request. The team must deliver the approved control safely. Which deployment sequence is most appropriate?
- Track duplicate request keys and erroneous retry policy changes
- Ship client-library changes with idempotency logging and gradual release
- Use idempotency keys and safe retry policy for mutation requests
- Enable unlimited blind retries on every POST
Explanation
Correct Answer
B:
Ship client-library changes with idempotency logging and gradual release is a controlled means to introduce the approved architecture while meeting the requirement that appointments must not be created twice for one patient request. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Track duplicate request keys and erroneous retry policy changes” addresses later monitoring rather than the actual controlled deployment. In the reported observation, trace shows repeated POST requests after an upstream timeout. The option “Ship client-library changes with idempotency logging and gradual release” instead addresses the requested controlled implementation.
C:
The alternative “Use idempotency keys and safe retry policy for mutation requests” describes the chosen design, but not how it is introduced safely. In the reported observation, trace shows repeated POST requests after an upstream timeout. The option “Ship client-library changes with idempotency logging and gradual release” instead addresses the requested controlled implementation.
D:
Unbounded retries can repeat side effects. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that appointments must not be created twice for one patient request.
Question 312
At Medical translation nonprofit, the Google Cloud design for speech-to-text triage is affected because clinics need transcripts of spoken patient questions and multilingual summaries. A review found that audio samples vary in background noise and language code switching. The proposed response must account for the requirement that patient data must be handled under consent and short retention. The cause has been established and a lasting architecture is being selected. Which design best meets the constraint?
- Monitor failed transcriptions, language bias and data-deletion completions
- Use an image-labeling API as the only audio transcriber
- Benchmark speech recognition error by language and noise against ground truth
- Use an appropriate speech recognition API with separate governed translation and summarization stages
Explanation
Correct Answer
D:
Use an appropriate speech recognition API with separate governed translation and summarization stages targets the underlying issue (clinics need transcripts of spoken patient questions and multilingual summaries) rather than merely reporting symptoms. It also preserves the required outcome (patient data must be handled under consent and short retention), which a target-state design must satisfy.
Incorrect Answers
A:
The alternative “Monitor failed transcriptions, language bias and data-deletion completions” measures results without defining the preventive design now required. In the reported observation, audio samples vary in background noise and language code switching. The option “Use an appropriate speech recognition API with separate governed translation and summarization stages” instead addresses the requested architecture selection.
B:
An image classification service does not provide speech transcripts. For this question, it is not a defensible choice for architecture selection and it does not satisfy the requirement that patient data must be handled under consent and short retention.
C:
The alternative “Benchmark speech recognition error by language and noise against ground truth” investigates the cause without specifying the durable architecture to resolve it. In the reported observation, audio samples vary in background noise and language code switching. The option “Use an appropriate speech recognition API with separate governed translation and summarization stages” instead addresses the requested architecture selection.
Question 313
At Financial consumer lender, the PCI data encryption separation workload has a documented risk: application developers can read a cardholder encryption key and production data. The review found that role review shows one service account can decrypt and administer Cloud KMS. The architecture must also ensure that PCI program requires separation of key administration and data processing. Which post-change test most reliably distinguishes a functioning control from a merely successful installation?
- Rotate permissions and test service continuity during KMS grant changes
- Map encryption/decrypt flows, identity grants and cardholder scope
- Set a shared spreadsheet password on exported plaintext cards
- Verify developers cannot decrypt card records or destroy keys
Explanation
Correct Answer
D:
Verify developers cannot decrypt card records or destroy keys is a direct acceptance test of the path implicated by role review shows one service account can decrypt and administer Cloud KMS. It checks that the original risk (application developers can read a cardholder encryption key and production data) has actually been controlled and tests that permitted operations still meet the requirement that PCI program requires separation of key administration and data processing.
Incorrect Answers
A:
The alternative “Rotate permissions and test service continuity during KMS grant changes” documents change execution but not its operational acceptance result. In the reported observation, role review shows one service account can decrypt and administer Cloud KMS. The option “Verify developers cannot decrypt card records or destroy keys” instead addresses the requested acceptance verification.
B:
The alternative “Map encryption/decrypt flows, identity grants and cardholder scope” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, role review shows one service account can decrypt and administer Cloud KMS. The option “Verify developers cannot decrypt card records or destroy keys” instead addresses the requested acceptance verification.
C:
A shared password does not meet secure key custody requirements. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that PCI program requires separation of key administration and data processing.
Question 314
Consider Public transportation authority and its organization location constraint workload. The issue is that test projects in an overseas region unexpectedly receive personal commuter data; the current observation is that resource inventory shows unrestricted region creation in folder projects. The team must ensure that all raw rider IDs must remain in the designated country. Which continuous monitoring approach best reveals recurrence or changing dependencies?
- Try deploying a prohibited regional resource and confirm denial
- Monitor location constraint exceptions and cross-region transfers
- Apply supported organization location restrictions and approved regional storage architecture
- Document a preferred region but leave cloud APIs unrestricted
Explanation
Correct Answer
B:
Monitor location constraint exceptions and cross-region transfers reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (test projects in an overseas region unexpectedly receive personal commuter data) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
The alternative “Try deploying a prohibited regional resource and confirm denial” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, resource inventory shows unrestricted region creation in folder projects. The option “Monitor location constraint exceptions and cross-region transfers” instead addresses the requested ongoing operational assurance.
C:
The alternative “Apply supported organization location restrictions and approved regional storage architecture” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, resource inventory shows unrestricted region creation in folder projects. The option “Monitor location constraint exceptions and cross-region transfers” instead addresses the requested ongoing operational assurance.
D:
A recommendation alone does not prevent prohibited region deployments. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that all raw rider IDs must remain in the designated country.
Question 315
At Healthcare AI supplier, the GKE node-upgrade safety workload has a documented risk: a node upgrade evicts all inference replicas because they share one failure domain. The review found that cluster events show two replicas pinned to the same maintenance pool. The architecture must also ensure that inference must remain available during security patching. What follow-up metric or alert will identify emerging failures after the initial acceptance test?
- Execute a controlled node drain and verify inference availability
- Monitor unavailable replicas, failed upgrades and scheduling constraints
- Set replica count to one and disable disruption checks
- Design multi-zone GKE scheduling with PodDisruptionBudget and appropriate replicas
Explanation
Correct Answer
B:
Monitor unavailable replicas, failed upgrades and scheduling constraints reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (a node upgrade evicts all inference replicas because they share one failure domain) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
The alternative “Execute a controlled node drain and verify inference availability” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, cluster events show two replicas pinned to the same maintenance pool. The option “Monitor unavailable replicas, failed upgrades and scheduling constraints” instead addresses the requested ongoing operational assurance.
C:
A single replica cannot tolerate a node drain. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that inference must remain available during security patching.
D:
The alternative “Design multi-zone GKE scheduling with PodDisruptionBudget and appropriate replicas” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, cluster events show two replicas pinned to the same maintenance pool. The option “Monitor unavailable replicas, failed upgrades and scheduling constraints” instead addresses the requested ongoing operational assurance.
Question 316
At Public education agency, the Cloud Bigtable emulator development workload has a documented risk: test suite uses expensive production Bigtable instance for every integration run. The review found that billing shows CI generates thousands of synthetic rows in live student datasets. The architecture must also ensure that tests must be repeatable without touching real student data. The team must deliver the approved control safely. Which deployment sequence is most appropriate?
- Use an isolated Bigtable emulator or appropriately scoped local test environment
- Configure CI to use emulator endpoint with synthetic tables
- Monitor environment variable drift and integration test failures
- Grant CI full production Bigtable admin to simplify setup
Explanation
Correct Answer
B:
Configure CI to use emulator endpoint with synthetic tables is a controlled means to introduce the approved architecture while meeting the requirement that tests must be repeatable without touching real student data. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Use an isolated Bigtable emulator or appropriately scoped local test environment” describes the chosen design, but not how it is introduced safely. In the reported observation, billing shows CI generates thousands of synthetic rows in live student datasets. The option “Configure CI to use emulator endpoint with synthetic tables” instead addresses the requested controlled implementation.
C:
The alternative “Monitor environment variable drift and integration test failures” addresses later monitoring rather than the actual controlled deployment. In the reported observation, billing shows CI generates thousands of synthetic rows in live student datasets. The option “Configure CI to use emulator endpoint with synthetic tables” instead addresses the requested controlled implementation.
D:
Production permissions undermine isolation and create destructive test risk. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that tests must be repeatable without touching real student data.
Question 317
Consider Metropolitan power utility and its SCADA historian migration workload. The issue is that historian collectors send proprietary UDP and engineers propose direct internet exposure; the current observation is that packet survey shows embedded devices without TLS or authentication. The team must ensure that plant control networks must remain isolated from public ingress. Which continuous monitoring approach best reveals recurrence or changing dependencies?
- Give cloud VMs public routes into the SCADA control VLAN
- Verify no inbound internet path reaches operational devices and all samples reconcile
- Design a controlled telemetry bridge through an industrial DMZ to cloud storage
- Track DMZ policy deviations and telemetry gaps
Explanation
Correct Answer
D:
Track DMZ policy deviations and telemetry gaps reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (historian collectors send proprietary UDP and engineers propose direct internet exposure) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
Direct internet access violates industrial segmentation. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that plant control networks must remain isolated from public ingress.
B:
The alternative “Verify no inbound internet path reaches operational devices and all samples reconcile” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, packet survey shows embedded devices without TLS or authentication. The option “Track DMZ policy deviations and telemetry gaps” instead addresses the requested ongoing operational assurance.
C:
The alternative “Design a controlled telemetry bridge through an industrial DMZ to cloud storage” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, packet survey shows embedded devices without TLS or authentication. The option “Track DMZ policy deviations and telemetry gaps” instead addresses the requested ongoing operational assurance.
Question 318
At Hospital payments administrator, the database point-in-time recovery workload has a documented risk: an operator accidentally ran a broad UPDATE against the claims settlement database. The review found that audit shows thousands of records changed before detection. The architecture must also ensure that finance requires restoration to immediately before the erroneous transaction. The team must deliver the approved control safely. Which deployment sequence is most appropriate?
- Restore from a week-old cold snapshot and overwrite new valid payments blindly
- Set continuous backup/PITR where supported and segregated recovery permissions
- Track log-retention windows, backup gaps and restore test age
- Enable recovery policy and rehearse isolated restore before relying on it
Explanation
Correct Answer
D:
Enable recovery policy and rehearse isolated restore before relying on it is a controlled means to introduce the approved architecture while meeting the requirement that finance requires restoration to immediately before the erroneous transaction. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
An old snapshot without transaction reconciliation can discard legitimate payments. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that finance requires restoration to immediately before the erroneous transaction.
B:
The alternative “Set continuous backup/PITR where supported and segregated recovery permissions” describes the chosen design, but not how it is introduced safely. In the reported observation, audit shows thousands of records changed before detection. The option “Enable recovery policy and rehearse isolated restore before relying on it” instead addresses the requested controlled implementation.
C:
The alternative “Track log-retention windows, backup gaps and restore test age” addresses later monitoring rather than the actual controlled deployment. In the reported observation, audit shows thousands of records changed before detection. The option “Enable recovery policy and rehearse isolated restore before relying on it” instead addresses the requested controlled implementation.
Question 319
Case study: National agricultural extension office: drought forecasts
National agricultural extension office relies on drought forecasts to deliver its operational commitments. The current architectural problem is that satellite datasets will multiply as climate partners join and current retention assumes fixed volume. The team has recorded this evidence: forecast identifies increasing seasonal data and changing external contracts. Any proposed response must preserve the requirement that field staff still need affordable offline reports.
The drought forecasts environment at National agricultural extension office faces this issue: satellite datasets will multiply as climate partners join and current retention assumes fixed volume. Supporting evidence shows that forecast identifies increasing seasonal data and changing external contracts. The architect must preserve the requirement that field staff still need affordable offline reports. What test must pass before the business accepts the remediated architecture?
- Migrate forecast periods progressively with access checks and archive-retrieval tests
- Model data retention, new partnership terms and user-access needs
- Keep every raw satellite capture indefinitely in the highest-price storage class
- Retrieve old seasonal maps and publish offline reports within the stated budget
Explanation
Correct Answer
D:
Retrieve old seasonal maps and publish offline reports within the stated budget is a direct acceptance test of the path implicated by forecast identifies increasing seasonal data and changing external contracts. It checks that the original risk (satellite datasets will multiply as climate partners join and current retention assumes fixed volume) has actually been controlled and tests that permitted operations still meet the requirement that field staff still need affordable offline reports.
Incorrect Answers
A:
The alternative “Migrate forecast periods progressively with access checks and archive-retrieval tests” documents change execution but not its operational acceptance result. In the reported observation, forecast identifies increasing seasonal data and changing external contracts. The option “Retrieve old seasonal maps and publish offline reports within the stated budget” instead addresses the requested acceptance verification.
B:
The alternative “Model data retention, new partnership terms and user-access needs” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, forecast identifies increasing seasonal data and changing external contracts. The option “Retrieve old seasonal maps and publish offline reports within the stated budget” instead addresses the requested acceptance verification.
C:
A fixed hot-storage policy wastes funds and ignores changing licenses. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that field staff still need affordable offline reports.
Question 320
At Multinational auto-insurer, the third-party risk boundary workload has a documented risk: a data processor claims SOC 2 coverage for a separate custom mobile collection app. The review found that contract review shows its audit excludes the in-app analytics SDK. The architecture must also ensure that customer consent and app data handling require independent assessment. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Map coverage of supplier attestations versus app and SDK control ownership
- Roll out SDK controls under contractual and app store change review
- Monitor third-party scope changes and mobile data egress
- Define separate privacy and security controls for the unassessed mobile collection path
- Treat the processor's SOC 2 report as certification of every unrelated mobile SDK
Explanation
Correct Answers
B:
Roll out SDK controls under contractual and app store change review is a controlled means to introduce the approved architecture while meeting the requirement that customer consent and app data handling require independent assessment. 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:
Define separate privacy and security controls for the unassessed mobile collection path targets the underlying issue (a data processor claims SOC 2 coverage for a separate custom mobile collection app) rather than merely reporting symptoms. It also preserves the required outcome (customer consent and app data handling require independent assessment), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
Incorrect Answers
A:
This diagnoses the original issue (a data processor claims SOC 2 coverage for a separate custom mobile collection app), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
C:
Monitoring third-party risk boundary for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: a data processor claims SOC 2 coverage for a separate custom mobile collection app.
E:
Attestation scope cannot be extended to unexamined components. The shortcut does not supply the requested design and safe rollout for third-party risk boundary, where contract review shows its audit excludes the in-app analytics SDK.