Google Professional Cloud Architect Practice Test 73
Question 720
The database engine transition environment at Public pension administrator faces this issue: an old relational engine uses vendor-specific stored procedures that block a simple migration. Supporting evidence shows that assessment finds financial calculation logic embedded in triggers. The architect must preserve the requirement that pension payouts require exact money arithmetic across historical backfills. Go-live and initial tests passed. Which ongoing measure is most useful to catch control drift or regression?
- Compare exact payout cents and historical audit attributes before cutover
- Plan staged conversion with verified query/procedure compatibility
- Treat stored procedures as portable SQL without regression tests
- Track procedure error rates and unauthorized schema changes
Explanation
Correct Answer
D:
Track procedure error rates and unauthorized schema changes reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (an old relational engine uses vendor-specific stored procedures that block a simple migration) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
The alternative “Compare exact payout cents and historical audit attributes before cutover” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, assessment finds financial calculation logic embedded in triggers. The option “Track procedure error rates and unauthorized schema changes” instead addresses the requested ongoing operational assurance.
B:
The alternative “Plan staged conversion with verified query/procedure compatibility” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, assessment finds financial calculation logic embedded in triggers. The option “Track procedure error rates and unauthorized schema changes” instead addresses the requested ongoing operational assurance.
C:
Vendor dialect and arithmetic differences can corrupt financial calculations. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that pension payouts require exact money arithmetic across historical backfills.
Question 721
Pharmaceutical research organization plans to update stakeholder release approvals. Investigators found that change journal shows data exports activated without regulatory signoff, consistent with the reported problem that engineering can launch clinical analytics while privacy and quality officers discover changes afterward. The response must satisfy the requirement that regulated data usage requires separate business and privacy approval. The security and reliability teams dispute the root cause. Which evidence-gathering action is justified now?
- Introduce approval gates in delivery workflow with emergency exception routing
- Let the fastest engineering manager approve all regulatory exceptions alone
- Design accountable change review with documented privacy/quality checkpoints
- Map decision rights, release gates and required stakeholder evidence
Explanation
Correct Answer
D:
Map decision rights, release gates and required stakeholder evidence is the appropriate diagnostic because the observation—change journal shows data exports activated without regulatory signoff—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (engineering can launch clinical analytics while privacy and quality officers discover changes afterward) before the design is approved.
Incorrect Answers
A:
The alternative “Introduce approval gates in delivery workflow with emergency exception routing” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, change journal shows data exports activated without regulatory signoff. The option “Map decision rights, release gates and required stakeholder evidence” instead addresses the requested root-cause evidence.
B:
Independent controls require the designated privacy/quality owners. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that regulated data usage requires separate business and privacy approval.
C:
The alternative “Design accountable change review with documented privacy/quality checkpoints” is target architecture selection; it does not establish why the observed path failed. In the reported observation, change journal shows data exports activated without regulatory signoff. The option “Map decision rights, release gates and required stakeholder evidence” instead addresses the requested root-cause evidence.
Question 722
At International hotel operator, the Google Cloud design for reservation mainframe bridge is affected because reservation jobs on a mainframe call undocumented nightly batch services. A review found that discovery captures eleven downstream dependencies absent from design diagrams. The proposed response must account for the requirement that bookings may not be lost during peak holiday weeks. Which controlled implementation should be used to preserve operational continuity during the change?
- Design strangler APIs around stable reservation capabilities with coexistence
- Track batch reconciliation, shadow traffic failures and residual dependencies
- Cut all mainframe links on the first deployment without inventory reconciliation
- Pilot low-risk read features and dual-write reconciliation before reservation cutover
Explanation
Correct Answer
D:
Pilot low-risk read features and dual-write reconciliation before reservation cutover is a controlled means to introduce the approved architecture while meeting the requirement that bookings may not be lost during peak holiday weeks. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Design strangler APIs around stable reservation capabilities with coexistence” describes the chosen design, but not how it is introduced safely. In the reported observation, discovery captures eleven downstream dependencies absent from design diagrams. The option “Pilot low-risk read features and dual-write reconciliation before reservation cutover” instead addresses the requested controlled implementation.
B:
The alternative “Track batch reconciliation, shadow traffic failures and residual dependencies” addresses later monitoring rather than the actual controlled deployment. In the reported observation, discovery captures eleven downstream dependencies absent from design diagrams. The option “Pilot low-risk read features and dual-write reconciliation before reservation cutover” instead addresses the requested controlled implementation.
C:
A big-bang cutover risks invisible dependencies and lost bookings. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that bookings may not be lost during peak holiday weeks.
Question 723
Regional airline booking platform plans to update Apigee API consumer management. Investigators found that gateway logs show one agency's retry loop overwhelms others, consistent with the reported problem that travel agencies call the same reservation endpoint with no consumer-specific quota. The response must satisfy the requirement that partners need predictable access without shutting down public bookings. Which target-state design is best justified by the evidence and required business outcome?
- Track per-client quota hits and anomalous retry behavior
- Identify API consumers, quota profiles and retry error causes
- Apply one secret shared by every travel agency with no rate controls
- Design Apigee API products with per-consumer credentials, quotas and analytics
Explanation
Correct Answer
D:
Design Apigee API products with per-consumer credentials, quotas and analytics targets the underlying issue (travel agencies call the same reservation endpoint with no consumer-specific quota) rather than merely reporting symptoms. It also preserves the required outcome (partners need predictable access without shutting down public bookings), which a target-state design must satisfy.
Incorrect Answers
A:
The alternative “Track per-client quota hits and anomalous retry behavior” measures results without defining the preventive design now required. In the reported observation, gateway logs show one agency's retry loop overwhelms others. The option “Design Apigee API products with per-consumer credentials, quotas and analytics” instead addresses the requested architecture selection.
B:
The alternative “Identify API consumers, quota profiles and retry error causes” investigates the cause without specifying the durable architecture to resolve it. In the reported observation, gateway logs show one agency's retry loop overwhelms others. The option “Design Apigee API products with per-consumer credentials, quotas and analytics” instead addresses the requested architecture selection.
C:
Shared identity and unrestricted rate make abusive tenants unaccountable. For this question, it is not a defensible choice for architecture selection and it does not satisfy the requirement that partners need predictable access without shutting down public bookings.
Question 724
Case study: Regional museum: global artifact web gallery
Regional museum relies on global artifact web gallery to deliver its operational commitments. The current architectural problem is that public image delivery overloads the origin service on exhibit announcement days. The team has recorded this evidence: traces show the same static media repeatedly retrieved by remote browsers. Any proposed response must preserve the requirement that private curator uploads must remain restricted while published assets are cacheable.
At Regional museum, the Google Cloud design for global artifact web gallery is affected because public image delivery overloads the origin service on exhibit announcement days. A review found that traces show the same static media repeatedly retrieved by remote browsers. The proposed response must account for the requirement that private curator uploads must remain restricted while published assets are cacheable. Which controlled implementation should be used to preserve operational continuity during the change?
- Cache authenticated curator endpoints for every visitor under a shared key
- Serve public assets through Cloud CDN while keeping private edit APIs uncached
- Monitor cache invalidations, origin egress and signed-access errors
- Launch edge caching for signed-public asset paths with invalidation procedure
Explanation
Correct Answer
D:
Launch edge caching for signed-public asset paths with invalidation procedure is a controlled means to introduce the approved architecture while meeting the requirement that private curator uploads must remain restricted while published assets are cacheable. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
Shared caching of private responses can leak unreleased content. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that private curator uploads must remain restricted while published assets are cacheable.
B:
The alternative “Serve public assets through Cloud CDN while keeping private edit APIs uncached” describes the chosen design, but not how it is introduced safely. In the reported observation, traces show the same static media repeatedly retrieved by remote browsers. The option “Launch edge caching for signed-public asset paths with invalidation procedure” instead addresses the requested controlled implementation.
C:
The alternative “Monitor cache invalidations, origin egress and signed-access errors” addresses later monitoring rather than the actual controlled deployment. In the reported observation, traces show the same static media repeatedly retrieved by remote browsers. The option “Launch edge caching for signed-public asset paths with invalidation procedure” instead addresses the requested controlled implementation.
Question 725
At Digital archive service, the Google Cloud design for object version recovery is affected because a script deleted storage objects and replication copied the deletions to the only backup. A review found that restore exercise shows no retained noncurrent version after replication. The proposed response must account for the requirement that a contractual RPO requires recovery from mistaken deletes for 45 days. Which analysis will allow the architect to distinguish the reported symptom from the actual failure path?
- Use versioning plus protected retention in a separately controlled recovery bucket
- Set lifecycle to delete every previous generation immediately
- Trace versioning, retention, replication and administrator delete permissions
- Enable retention after validating governance authority and object lifecycle effects
Explanation
Correct Answer
C:
Trace versioning, retention, replication and administrator delete permissions is the appropriate diagnostic because the observation—restore exercise shows no retained noncurrent version after replication—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (a script deleted storage objects and replication copied the deletions to the only backup) before the design is approved.
Incorrect Answers
A:
The alternative “Use versioning plus protected retention in a separately controlled recovery bucket” is target architecture selection; it does not establish why the observed path failed. In the reported observation, restore exercise shows no retained noncurrent version after replication. The option “Trace versioning, retention, replication and administrator delete permissions” instead addresses the requested root-cause evidence.
B:
Immediate deletion removes the very recovery history required. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that a contractual RPO requires recovery from mistaken deletes for 45 days.
D:
The alternative “Enable retention after validating governance authority and object lifecycle effects” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, restore exercise shows no retained noncurrent version after replication. The option “Trace versioning, retention, replication and administrator delete permissions” instead addresses the requested root-cause evidence.
Question 726
Community public library plans to update Terraform landing zone rollout. Investigators found that audit finds mutable local state and unreviewed IAM grants, consistent with the reported problem that new projects require approved security settings but Terraform changes are applied directly from laptops. The response must satisfy the requirement that public catalog availability must not be interrupted by infrastructure drift. The approved change has been applied. Which validation provides the clearest evidence the original problem is fixed?
- Map infrastructure state, approval roles and module dependencies
- Apply reviewed Terraform changes by staged projects with rollback
- Give all engineers full project Owner for local Terraform apply
- Verify identical approved plan outcomes and no unreviewed IAM escalation
Explanation
Correct Answer
D:
Verify identical approved plan outcomes and no unreviewed IAM escalation is a direct acceptance test of the path implicated by audit finds mutable local state and unreviewed IAM grants. It checks that the original risk (new projects require approved security settings but Terraform changes are applied directly from laptops) has actually been controlled and tests that permitted operations still meet the requirement that public catalog availability must not be interrupted by infrastructure drift.
Incorrect Answers
A:
The alternative “Map infrastructure state, approval roles and module dependencies” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, audit finds mutable local state and unreviewed IAM grants. The option “Verify identical approved plan outcomes and no unreviewed IAM escalation” instead addresses the requested acceptance verification.
B:
The alternative “Apply reviewed Terraform changes by staged projects with rollback” documents change execution but not its operational acceptance result. In the reported observation, audit finds mutable local state and unreviewed IAM grants. The option “Verify identical approved plan outcomes and no unreviewed IAM escalation” instead addresses the requested acceptance verification.
C:
Unreviewed broad changes negate controlled infrastructure rollout. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that public catalog availability must not be interrupted by infrastructure drift.
Question 727
Case study: Regional transport network: low-latency dispatch API
Regional transport network relies on low-latency dispatch API to deliver its operational commitments. The current architectural problem is that dispatch requests cross the internet to an API hosted in another region. The team has recorded this evidence: flow logs show round trips go through a centralized NAT despite local backends. Any proposed response must preserve the requirement that private dispatch callers require restricted routing and prompt responses.
At Regional transport network, the low-latency dispatch API workload has a documented risk: dispatch requests cross the internet to an API hosted in another region. The review found that flow logs show round trips go through a centralized NAT despite local backends. The architecture must also ensure that private dispatch callers require restricted routing and prompt responses. The architecture review is paused until the team can prove the real failure path. Which investigation should happen first?
- Choose regional internal load balancing and private connectivity paths
- Inspect hybrid route tables, peering limitations and service DNS
- Put a public forwarding rule on every internal backend for simplicity
- Gradually switch dispatch clients with route validation and rollback
Explanation
Correct Answer
B:
Inspect hybrid route tables, peering limitations and service DNS is the appropriate diagnostic because the observation—flow logs show round trips go through a centralized NAT despite local backends—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (dispatch requests cross the internet to an API hosted in another region) before the design is approved.
Incorrect Answers
A:
The alternative “Choose regional internal load balancing and private connectivity paths” is target architecture selection; it does not establish why the observed path failed. In the reported observation, flow logs show round trips go through a centralized NAT despite local backends. The option “Inspect hybrid route tables, peering limitations and service DNS” instead addresses the requested root-cause evidence.
C:
Public ingress violates the explicit private dispatch boundary. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that private dispatch callers require restricted routing and prompt responses.
D:
The alternative “Gradually switch dispatch clients with route validation and rollback” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, flow logs show round trips go through a centralized NAT despite local backends. The option “Inspect hybrid route tables, peering limitations and service DNS” instead addresses the requested root-cause evidence.
Question 728
At Retail financial platform, the edge web application defense workload has a documented risk: automated scraping exhausts the app before request processing while legitimate checkout continues. The review found that HTTP logs show one ASN generates most high-rate suspicious requests. The architecture must also ensure that payments require low false-positive blocking and reliable low-latency access. What follow-up metric or alert will identify emerging failures after the initial acceptance test?
- Apply Cloud Armor policies to the external HTTPS load balancer with adaptive protections
- Run benign and abusive traffic tests through the load balancer and compare outcomes
- Rely only on a VM host firewall to examine full application-layer request behavior
- Watch false-positive blocks, attack throughput and backend saturation
Explanation
Correct Answer
D:
Watch false-positive blocks, attack throughput and backend saturation reveals recurrence, drift or changing dependencies after launch. It provides a repeatable signal for the risk (automated scraping exhausts the app before request processing while legitimate checkout continues) rather than treating the first successful test as permanent assurance.
Incorrect Answers
A:
The alternative “Apply Cloud Armor policies to the external HTTPS load balancer with adaptive protections” selects a design, but does not provide the ongoing signal required to detect drift. In the reported observation, HTTP logs show one ASN generates most high-rate suspicious requests. The option “Watch false-positive blocks, attack throughput and backend saturation” instead addresses the requested ongoing operational assurance.
B:
The alternative “Run benign and abusive traffic tests through the load balancer and compare outcomes” is a one-time acceptance result rather than ongoing operational assurance. In the reported observation, HTTP logs show one ASN generates most high-rate suspicious requests. The option “Watch false-positive blocks, attack throughput and backend saturation” instead addresses the requested ongoing operational assurance.
C:
Host-level rules do not provide the same edge application-aware filtering. For this question, it is not a defensible choice for ongoing operational assurance and it does not satisfy the requirement that payments require low false-positive blocking and reliable low-latency access.
Question 729
At Regional vehicle-parts supplier, the Google Cloud design for hybrid stock synchronization is affected because headquarters reaches Google Cloud only through one IPsec tunnel that drops during maintenance. A review found that network log shows packets re-route onto a congested internet path after the gateway reboot. The proposed response must account for the requirement that inventory updates need redundant private routing to two regions. A deployment ticket has been closed. What direct evidence would justify solution signoff?
- Establish both encrypted paths and verify advertised routes before removing the old tunnel
- Inspect tunnel failure modes, BGP advertisements and required private destinations
- Fail one peer and prove private inventory replication stays available
- Use a single static VPN gateway and advertise a default public route for all stock traffic
Explanation
Correct Answer
C:
Fail one peer and prove private inventory replication stays available is a direct acceptance test of the path implicated by network log shows packets re-route onto a congested internet path after the gateway reboot. It checks that the original risk (headquarters reaches Google Cloud only through one IPsec tunnel that drops during maintenance) has actually been controlled and tests that permitted operations still meet the requirement that inventory updates need redundant private routing to two regions.
Incorrect Answers
A:
The alternative “Establish both encrypted paths and verify advertised routes before removing the old tunnel” documents change execution but not its operational acceptance result. In the reported observation, network log shows packets re-route onto a congested internet path after the gateway reboot. The option “Fail one peer and prove private inventory replication stays available” instead addresses the requested acceptance verification.
B:
The alternative “Inspect tunnel failure modes, BGP advertisements and required private destinations” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, network log shows packets re-route onto a congested internet path after the gateway reboot. The option “Fail one peer and prove private inventory replication stays available” instead addresses the requested acceptance verification.
D:
One tunnel and unscoped route reproduce the outage and security risk. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that inventory updates need redundant private routing to two regions.
Question 730
At International digital learning network, the Cloud Trace bottleneck workload has a documented risk: users see slow lesson page load but service-level metrics show each component as healthy. The review found that trace spans reveal sequential requests to a distant analytics service. The architecture must also ensure that lesson pages must remain responsive if nonessential analytics pauses. The team must deliver the approved control safely. Which deployment sequence is most appropriate?
- Increase every VM size without checking span-level critical path
- Introduce asynchronous recording and gradually shift user cohorts
- Move optional analytics off the synchronous critical path with trace instrumentation
- Watch trace critical-path latency and queued analytics backlog
Explanation
Correct Answer
B:
Introduce asynchronous recording and gradually shift user cohorts is a controlled means to introduce the approved architecture while meeting the requirement that lesson pages must remain responsive if nonessential analytics pauses. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
Generic scaling need not solve avoidable synchronous waiting. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that lesson pages must remain responsive if nonessential analytics pauses.
C:
The alternative “Move optional analytics off the synchronous critical path with trace instrumentation” describes the chosen design, but not how it is introduced safely. In the reported observation, trace spans reveal sequential requests to a distant analytics service. The option “Introduce asynchronous recording and gradually shift user cohorts” instead addresses the requested controlled implementation.
D:
The alternative “Watch trace critical-path latency and queued analytics backlog” addresses later monitoring rather than the actual controlled deployment. In the reported observation, trace spans reveal sequential requests to a distant analytics service. The option “Introduce asynchronous recording and gradually shift user cohorts” instead addresses the requested controlled implementation.