Google Professional Cloud Architect Practice Test 61
Question 600
Consider City parking payment app and its payment token audit workload. The issue is that service records full PAN in debug logs while application database is tokenized; the current observation is that log sample includes unmasked card numbers in error traces. The team must ensure that payment operations require minimized log retention and incident-ready evidence. Which continuous monitoring approach best reveals recurrence or changing dependencies?
- Monitor new PAN detections and changes in logging pipeline
- Use TLS on log transport and keep full card numbers indefinitely
- Remove PAN from telemetry and apply scoped tokenization and log redaction
- Generate test errors and verify no PAN appears in log destinations
Explanation
Correct Answer
A:
Monitor new PAN detections and changes in logging pipeline reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (service records full PAN in debug logs while application database is tokenized) rather than treating the first successful test as permanent assurance.
Incorrect Answers
B:
Transport encryption alone does not minimize sensitive logged content. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that payment operations require minimized log retention and incident-ready evidence.
C:
The alternative “Remove PAN from telemetry and apply scoped tokenization and log redaction” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, log sample includes unmasked card numbers in error traces. The option “Monitor new PAN detections and changes in logging pipeline” instead addresses the requested ongoing operational assurance.
D:
The alternative “Generate test errors and verify no PAN appears in log destinations” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, log sample includes unmasked card numbers in error traces. The option “Monitor new PAN detections and changes in logging pipeline” instead addresses the requested ongoing operational assurance.
Question 601
Consider Digital university archive and its BigQuery scripted exports workload. The issue is that a nightly bq CLI export uses wildcard datasets and leaks unrelated faculty research; the current observation is that audit log shows query scope expanded after a new dataset was added. The team must ensure that exports must include only approved course statistics. The target architecture is approved; execution without disruption is the remaining decision. Which rollout action is best?
- Use parameterized explicit BigQuery dataset/table scope with read-only job identity
- Give the export service account BigQuery Admin over the whole organization
- Deploy export script with bounded project context and approved schedules
- Track permission changes and unexpected query bytes
Explanation
Correct Answer
C:
Deploy export script with bounded project context and approved schedules is a controlled means to introduce the approved architecture while meeting the requirement that exports must include only approved course statistics. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Use parameterized explicit BigQuery dataset/table scope with read-only job identity” describes the chosen design, but not how it is introduced safely. In the reported observation, audit log shows query scope expanded after a new dataset was added. The option “Deploy export script with bounded project context and approved schedules” instead addresses the requested controlled implementation.
B:
Broad admin rights violate the limited read-only extraction purpose. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that exports must include only approved course statistics.
D:
The alternative “Track permission changes and unexpected query bytes” addresses later monitoring rather than the actual controlled deployment. In the reported observation, audit log shows query scope expanded after a new dataset was added. The option “Deploy export script with bounded project context and approved schedules” instead addresses the requested controlled implementation.
Question 602
At Regional financial adviser, the Google Cloud design for private research assistant is affected because employees want document-grounded answers without granting a chatbot access to privileged trading APIs. A review found that demo agent follows instructions embedded in retrieved research PDFs. The proposed response must account for the requirement that approved internal reports must be cited and trading authorization separate. Which analysis will allow the architect to distinguish the reported symptom from the actual failure path?
- Ground answers in vetted enterprise search and isolate actions behind explicit identity checks
- Treat every retrieved document as an instruction from the security administrator
- Introduce approved corpus sources with prompt-injection tests and read-only permissions
- Map trusted retrieved context and the tool-authorization boundary
Explanation
Correct Answer
D:
Map trusted retrieved context and the tool-authorization boundary is the appropriate diagnostic because the observation—demo agent follows instructions embedded in retrieved research PDFs—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (employees want document-grounded answers without granting a chatbot access to privileged trading APIs) before the design is approved.
Incorrect Answers
A:
The alternative “Ground answers in vetted enterprise search and isolate actions behind explicit identity checks” is target architecture selection; it does not establish why the observed path failed. In the reported observation, demo agent follows instructions embedded in retrieved research PDFs. The option “Map trusted retrieved context and the tool-authorization boundary” instead addresses the requested root-cause evidence.
B:
Untrusted document content cannot grant privileged tool authority. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that approved internal reports must be cited and trading authorization separate.
C:
The alternative “Introduce approved corpus sources with prompt-injection tests and read-only permissions” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, demo agent follows instructions embedded in retrieved research PDFs. The option “Map trusted retrieved context and the tool-authorization boundary” instead addresses the requested root-cause evidence.
Question 603
At Airport luggage services, the service catalog onboarding workload has a documented risk: new teams ask for databases by email and receive unrestricted project ownership. The review found that tickets show inconsistent machine sizing and missing application tags. The architecture must also ensure that every project needs approved cost and access boundaries. Which post-change test most reliably distinguishes a functioning control from a merely successful installation?
- Launch approved Terraform-backed catalog items for pilot teams
- Continue granting project Owner for all simple database requests
- Provision test databases and verify limits, tagging and privileges
- Review common database requests, service tiers and authorized provisioning roles
Explanation
Correct Answer
C:
Provision test databases and verify limits, tagging and privileges is a direct acceptance test of the path implicated by tickets show inconsistent machine sizing and missing application tags. It checks that the original risk (new teams ask for databases by email and receive unrestricted project ownership) has actually been controlled and tests that permitted operations still meet the requirement that every project needs approved cost and access boundaries.
Incorrect Answers
A:
The alternative “Launch approved Terraform-backed catalog items for pilot teams” documents change execution but not its operational acceptance result. In the reported observation, tickets show inconsistent machine sizing and missing application tags. The option “Provision test databases and verify limits, tagging and privileges” instead addresses the requested acceptance verification.
B:
Broad manual grants defeat governed repeatability. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that every project needs approved cost and access boundaries.
D:
The alternative “Review common database requests, service tiers and authorized provisioning roles” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, tickets show inconsistent machine sizing and missing application tags. The option “Provision test databases and verify limits, tagging and privileges” instead addresses the requested acceptance verification.
Question 604
Consider Online grocery delivery and its feature KPI design workload. The issue is that migration success is defined as number of VMs shut down while delivery failures increase; the current observation is that business dashboards show missed deliveries after new routing service launch. The team must ensure that the program must improve delivered-order reliability, not just infrastructure count. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Reconcile technical metrics with customer outcomes and fulfillment costs
- Track on-time rate, order cancellations and unit economics
- Define shared availability, on-time-delivery and cost-per-delivery KPIs
- Call the project successful solely because the cloud invoice fell
- Introduce metrics dashboards with business owners and stable baselines
Explanation
Correct Answers
C:
Define shared availability, on-time-delivery and cost-per-delivery KPIs targets the underlying issue (migration success is defined as number of VMs shut down while delivery failures increase) rather than merely reporting symptoms. It also preserves the required outcome (the program must improve delivered-order reliability, not just infrastructure count), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
E:
Introduce metrics dashboards with business owners and stable baselines is a controlled means to introduce the approved architecture while meeting the requirement that the program must improve delivered-order reliability, not just infrastructure count. 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.
Incorrect Answers
A:
This diagnoses the original issue (migration success is defined as number of VMs shut down while delivery failures increase), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
B:
Monitoring feature KPI design for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: migration success is defined as number of VMs shut down while delivery failures increase.
D:
Lower spend can accompany worse business performance. The shortcut does not supply the requested design and safe rollout for feature KPI design, where business dashboards show missed deliveries after new routing service launch.
Question 605
Sustainable packaging manufacturer plans to update energy-aware workload economics. Investigators found that billing data shows batch energy and instance costs spike in fixed early hours, consistent with the reported problem that nightly noncritical reports run during peak electricity pricing even though deadline is the next afternoon. The response must satisfy the requirement that reports must be complete by 14:00, not in real time. The security and reliability teams dispute the root cause. Which evidence-gathering action is justified now?
- Compare time flexibility, consumption profiles and data freshness
- Terminate all reports to maximize energy savings
- Trial alternative schedules with rollback before peak season
- Schedule nonurgent batch jobs in lower-cost windows subject to completion SLO
Explanation
Correct Answer
A:
Compare time flexibility, consumption profiles and data freshness is the appropriate diagnostic because the observation—billing data shows batch energy and instance costs spike in fixed early hours—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (nightly noncritical reports run during peak electricity pricing even though deadline is the next afternoon) before the design is approved.
Incorrect Answers
B:
Zero processing fails the defined reporting objective. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that reports must be complete by 14:00, not in real time.
C:
The alternative “Trial alternative schedules with rollback before peak season” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, billing data shows batch energy and instance costs spike in fixed early hours. The option “Compare time flexibility, consumption profiles and data freshness” instead addresses the requested root-cause evidence.
D:
The alternative “Schedule nonurgent batch jobs in lower-cost windows subject to completion SLO” is target architecture selection; it does not establish why the observed path failed. In the reported observation, billing data shows batch energy and instance costs spike in fixed early hours. The option “Compare time flexibility, consumption profiles and data freshness” instead addresses the requested root-cause evidence.
Question 606
Consider E-commerce promotion service and its Cloud Run database connection storms workload. The issue is that a serverless revision scales quickly and exhausts a legacy SQL server's connection pool; the current observation is that database logs show connection refusals at promotion launch. The team must ensure that backend must remain within database concurrency limits. Which acceptance test should demonstrate both protection against the failure and continued authorized service?
- Stage maximum-instance limits and traffic splitting before peak events
- Allow unlimited cold-start instances to open independent large pools
- Calculate per-instance connection pools, maximum instances and request bursts
- Generate realistic parallel traffic and verify connections stay within ceiling
Explanation
Correct Answer
D:
Generate realistic parallel traffic and verify connections stay within ceiling is a direct acceptance test of the path implicated by database logs show connection refusals at promotion launch. It checks that the original risk (a serverless revision scales quickly and exhausts a legacy SQL server's connection pool) has actually been controlled and tests that permitted operations still meet the requirement that backend must remain within database concurrency limits.
Incorrect Answers
A:
The alternative “Stage maximum-instance limits and traffic splitting before peak events” documents change execution but not its operational acceptance result. In the reported observation, database logs show connection refusals at promotion launch. The option “Generate realistic parallel traffic and verify connections stay within ceiling” instead addresses the requested acceptance verification.
B:
Unbounded scaling overwhelms a fixed database dependency. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that backend must remain within database concurrency limits.
C:
The alternative “Calculate per-instance connection pools, maximum instances and request bursts” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, database logs show connection refusals at promotion launch. The option “Generate realistic parallel traffic and verify connections stay within ceiling” instead addresses the requested acceptance verification.
Question 607
The shipment tracking platform environment at International shipping brokerage faces this issue: executives want real-time tracking without defining the latency and availability targets. Supporting evidence shows that port partners deliver events with variable delays and occasional duplicate identifiers. The architect must preserve the requirement that customs declarations require nonduplicated status transitions. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Architect idempotent event ingestion and a separately materialized customer tracking view
- Onboard one shipping partner with replayable event streams and reconciliation
- Agree on user-visible freshness, event semantics and declaration correctness with partners
- Measure lag per partner, invalid transitions and delivery-cost trends
- Assume ordered, exactly-once delivery from every external partner without validation
Explanation
Correct Answers
A:
Architect idempotent event ingestion and a separately materialized customer tracking view targets the underlying issue (executives want real-time tracking without defining the latency and availability targets) rather than merely reporting symptoms. It also preserves the required outcome (customs declarations require nonduplicated status transitions), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
B:
Onboard one shipping partner with replayable event streams and reconciliation is a controlled means to introduce the approved architecture while meeting the requirement that customs declarations require nonduplicated status transitions. 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.
Incorrect Answers
C:
This diagnoses the original issue (executives want real-time tracking without defining the latency and availability targets), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
D:
Monitoring shipment tracking platform for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: executives want real-time tracking without defining the latency and availability targets.
E:
External partner delivery can be duplicated or out of order. The shortcut does not supply the requested design and safe rollout for shipment tracking platform, where port partners deliver events with variable delays and occasional duplicate identifiers.
Question 608
At Telecom billing department, the patchable application instances workload has a documented risk: operators manually patch live VMs and drift from the declared golden configuration. The review found that audit finds two instances missing a required base image update. The architecture must also ensure that customers must not lose billing access during security patch cycles. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Compare instance metadata, patch windows and image drift
- SSH into all production servers simultaneously and update without rollback
- Adopt immutable image templates with rolling managed-instance replacements
- Track patch-level drift and failed rolling update cohorts
- Deploy template versions in health-gated batches across zones
Explanation
Correct Answers
C:
Adopt immutable image templates with rolling managed-instance replacements targets the underlying issue (operators manually patch live VMs and drift from the declared golden configuration) rather than merely reporting symptoms. It also preserves the required outcome (customers must not lose billing access during security patch cycles), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
E:
Deploy template versions in health-gated batches across zones is a controlled means to introduce the approved architecture while meeting the requirement that customers must not lose billing access during security patch cycles. 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.
Incorrect Answers
A:
This diagnoses the original issue (operators manually patch live VMs and drift from the declared golden configuration), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
B:
Uncoordinated in-place patching creates inconsistent state and outage risk. The shortcut does not supply the requested design and safe rollout for patchable application instances, where audit finds two instances missing a required base image update.
D:
Monitoring patchable application instances for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: operators manually patch live VMs and drift from the declared golden configuration.
Question 609
Consider Climate modeling consortium and its GPU inference profiling workload. The issue is that AI inference costs spike although request counts stay flat; the current observation is that profile shows a new preprocessing step copies tensors repeatedly between CPU and GPU. The team must ensure that cost optimization must not lower model output quality. The target architecture is approved; execution without disruption is the remaining decision. Which rollout action is best?
- Buy more GPUs without measuring the memory-transfer bottleneck
- Monitor transfer bandwidth, idle accelerator time and model drift
- Roll out optimized preprocessing behind a versioned serving endpoint
- Reduce redundant tensor transfer and right-size accelerators based on measured bottleneck
Explanation
Correct Answer
C:
Roll out optimized preprocessing behind a versioned serving endpoint is a controlled means to introduce the approved architecture while meeting the requirement that cost optimization must not lower model output quality. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
More compute may increase spend while copies still dominate. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that cost optimization must not lower model output quality.
B:
The alternative “Monitor transfer bandwidth, idle accelerator time and model drift” addresses later monitoring rather than the actual controlled deployment. In the reported observation, profile shows a new preprocessing step copies tensors repeatedly between CPU and GPU. The option “Roll out optimized preprocessing behind a versioned serving endpoint” instead addresses the requested controlled implementation.
D:
The alternative “Reduce redundant tensor transfer and right-size accelerators based on measured bottleneck” describes the chosen design, but not how it is introduced safely. In the reported observation, profile shows a new preprocessing step copies tensors repeatedly between CPU and GPU. The option “Roll out optimized preprocessing behind a versioned serving endpoint” instead addresses the requested controlled implementation.
Question 610
Case study: University HPC consortium: interruptible research training
University HPC consortium relies on interruptible research training to deliver its operational commitments. The current architectural problem is that lab runs preemptible VMs for long training without checkpoints and loses weeks of progress. The team has recorded this evidence: job logs show instances interrupted after capacity reclaim. Any proposed response must preserve the requirement that researcher budget is constrained but results must be reproducible.
Consider University HPC consortium and its interruptible research training workload. The issue is that lab runs preemptible VMs for long training without checkpoints and loses weeks of progress; the current observation is that job logs show instances interrupted after capacity reclaim. The team must ensure that researcher budget is constrained but results must be reproducible. Which continuous monitoring approach best reveals recurrence or changing dependencies?
- Use Spot VMs for restartable segments with durable checkpoints
- Interrupt a worker and compare resumed metrics with expected state
- Run the authoritative metadata database only on unreplicated Spot VMs
- Monitor preemption loss, checkpoint staleness and cost per experiment
Explanation
Correct Answer
D:
Monitor preemption loss, checkpoint staleness and cost per experiment reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (lab runs preemptible VMs for long training without checkpoints and loses weeks of progress) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
The alternative “Use Spot VMs for restartable segments with durable checkpoints” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, job logs show instances interrupted after capacity reclaim. The option “Monitor preemption loss, checkpoint staleness and cost per experiment” instead addresses the requested ongoing operational assurance.
B:
The alternative “Interrupt a worker and compare resumed metrics with expected state” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, job logs show instances interrupted after capacity reclaim. The option “Monitor preemption loss, checkpoint staleness and cost per experiment” instead addresses the requested ongoing operational assurance.
C:
The database cannot tolerate unmanaged interruption and state loss. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that researcher budget is constrained but results must be reproducible.