Google Professional Cloud Architect Practice Test 37
Question 360
At Multinational consulting alliance, the shared IAM inheritance workload has a documented risk: a merger added interns to a group that has editor permission on financial-production resources. The review found that IAM policy audit shows project roles inherited from the organization root. The architecture must also ensure that production access must be limited to verified operational duties. What enduring Google Cloud design should the team choose now that the failure mode is understood?
- Create a new group but keep the old wildcard organization binding
- Watch group membership changes and inherited privilege escalation
- Move broad grants to scoped groups and use conditional role bindings where appropriate
- Identify inherited member groups and smallest needed job privileges
Explanation
Correct Answer
C:
Move broad grants to scoped groups and use conditional role bindings where appropriate targets the underlying issue (a merger added interns to a group that has editor permission on financial-production resources) rather than merely reporting symptoms. It also preserves the required outcome (production access must be limited to verified operational duties), which a target-state design must satisfy.
Incorrect Answers
A:
A new group does not negate an existing broad inherited permission. For this question, it is not a defensible choice for architecture selection and it does not satisfy the requirement that production access must be limited to verified operational duties.
B:
The alternative “Watch group membership changes and inherited privilege escalation” measures results without defining the preventive design now required. In the reported observation, IAM policy audit shows project roles inherited from the organization root. The option “Move broad grants to scoped groups and use conditional role bindings where appropriate” instead addresses the requested architecture selection.
D:
The alternative “Identify inherited member groups and smallest needed job privileges” investigates the cause without specifying the durable architecture to resolve it. In the reported observation, IAM policy audit shows project roles inherited from the organization root. The option “Move broad grants to scoped groups and use conditional role bindings where appropriate” instead addresses the requested architecture selection.
Question 361
National electricity retailer plans to update cloud vendor contract renewal. Investigators found that network bill shows cross-region replication costs rise each quarter, consistent with the reported problem that finance only compares monthly cloud spend and ignores data-transfer commitments. The response must satisfy the requirement that customer billing must meet a recovery objective and fixed budget. The design review is over and production introduction is next. Which action is the best fit?
- Disable all cross-region recovery copies solely to eliminate egress fees
- Implement cost guardrails after recovery testing and owner approval
- Choose a right-sized retention and replication strategy linked to RPO
- Monitor interregion egress and utilization of commitments
Explanation
Correct Answer
B:
Implement cost guardrails after recovery testing and owner approval is a controlled means to introduce the approved architecture while meeting the requirement that customer billing must meet a recovery objective and fixed budget. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
Deleting recovery ignores the stated continuity obligation. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that customer billing must meet a recovery objective and fixed budget.
C:
The alternative “Choose a right-sized retention and replication strategy linked to RPO” describes the chosen design, but not how it is introduced safely. In the reported observation, network bill shows cross-region replication costs rise each quarter. The option “Implement cost guardrails after recovery testing and owner approval” instead addresses the requested controlled implementation.
D:
The alternative “Monitor interregion egress and utilization of commitments” addresses later monitoring rather than the actual controlled deployment. In the reported observation, network bill shows cross-region replication costs rise each quarter. The option “Implement cost guardrails after recovery testing and owner approval” instead addresses the requested controlled implementation.
Question 362
Consider National property insurer and its claims analytics modernization workload. The issue is that finance compared storage prices without accounting for nightly claims integration and legal holds; the current observation is that workload inventory shows a batch export consumed by three regulators and an external actuary. The team must ensure that actuarial outputs must be delivered by 06:00 every business day. The team has completed discovery. Which preventive or corrective architecture is most appropriate?
- Define decoupled ingestion, governed analytical storage and a bounded overnight processing objective
- Trace upstream claims feeds, regulator deadlines, legal-hold periods and total run costs
- Copy all claim tables into public object buckets for analysts
- Monitor missed ingestion deadlines, unclassified data and projected storage spend
Explanation
Correct Answer
A:
Define decoupled ingestion, governed analytical storage and a bounded overnight processing objective targets the underlying issue (finance compared storage prices without accounting for nightly claims integration and legal holds) rather than merely reporting symptoms. It also preserves the required outcome (actuarial outputs must be delivered by 06:00 every business day), which a target-state design must satisfy.
Incorrect Answers
B:
The alternative “Trace upstream claims feeds, regulator deadlines, legal-hold periods and total run costs” investigates the cause without specifying the durable architecture to resolve it. In the reported observation, workload inventory shows a batch export consumed by three regulators and an external actuary. The option “Define decoupled ingestion, governed analytical storage and a bounded overnight processing objective” instead addresses the requested architecture selection.
C:
Public storage would violate claims confidentiality and regulated access. For this question, it is not a defensible choice for architecture selection and it does not satisfy the requirement that actuarial outputs must be delivered by 06:00 every business day.
D:
The alternative “Monitor missed ingestion deadlines, unclassified data and projected storage spend” measures results without defining the preventive design now required. In the reported observation, workload inventory shows a batch export consumed by three regulators and an external actuary. The option “Define decoupled ingestion, governed analytical storage and a bounded overnight processing objective” instead addresses the requested architecture selection.
Question 363
Municipal parking authority plans to update legacy Oracle billing. Investigators found that license assessment shows processor-based fees exceed projected cloud infrastructure savings, consistent with the reported problem that a tightly coupled database license cannot run cost-effectively in a naive VM lift-and-shift. The response must satisfy the requirement that metered invoices must reconcile across cutover. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Ignore license rights because compute is billed hourly
- Monitor license exposure, query regressions and reconciliation failures
- Select a phased database disposition using compatibility and total cost, not VM price alone
- Inventory dependencies, license terms, SQL compatibility and transaction owners
- Trial a compatible subset with dual invoices and a rollback date
Explanation
Correct Answers
C:
Select a phased database disposition using compatibility and total cost, not VM price alone targets the underlying issue (a tightly coupled database license cannot run cost-effectively in a naive VM lift-and-shift) rather than merely reporting symptoms. It also preserves the required outcome (metered invoices must reconcile across cutover), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
E:
Trial a compatible subset with dual invoices and a rollback date is a controlled means to introduce the approved architecture while meeting the requirement that metered invoices must reconcile across cutover. 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:
Cloud billing does not nullify software licensing obligations. The shortcut does not supply the requested design and safe rollout for legacy Oracle billing, where license assessment shows processor-based fees exceed projected cloud infrastructure savings.
B:
Monitoring legacy Oracle billing for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: a tightly coupled database license cannot run cost-effectively in a naive VM lift-and-shift.
D:
This diagnoses the original issue (a tightly coupled database license cannot run cost-effectively in a naive VM lift-and-shift), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
Question 364
Consider Global pharmacy vendor and its private AI endpoint access workload. The issue is that public model endpoints are reachable by unauthorized IPs while containing protected pricing prompts; the current observation is that network scan finds broad ingress on a model-serving load balancer. The team must ensure that only corporate approved callers may submit sensitive requests. The team has completed discovery. Which preventive or corrective architecture is most appropriate?
- Map service callers, endpoint type and private reachability needs
- Use private model-serving access and workload-bound IAM with constrained ingress
- Monitor ingress drift and attempted model endpoint abuse
- Hide endpoint names in source code but leave public access open
Explanation
Correct Answer
B:
Use private model-serving access and workload-bound IAM with constrained ingress targets the underlying issue (public model endpoints are reachable by unauthorized IPs while containing protected pricing prompts) rather than merely reporting symptoms. It also preserves the required outcome (only corporate approved callers may submit sensitive requests), which a target-state design must satisfy.
Incorrect Answers
A:
The alternative “Map service callers, endpoint type and private reachability needs” investigates the cause without specifying the durable architecture to resolve it. In the reported observation, network scan finds broad ingress on a model-serving load balancer. The option “Use private model-serving access and workload-bound IAM with constrained ingress” instead addresses the requested architecture selection.
C:
The alternative “Monitor ingress drift and attempted model endpoint abuse” measures results without defining the preventive design now required. In the reported observation, network scan finds broad ingress on a model-serving load balancer. The option “Use private model-serving access and workload-bound IAM with constrained ingress” instead addresses the requested architecture selection.
D:
Obscurity is not an endpoint authorization control. For this question, it is not a defensible choice for architecture selection and it does not satisfy the requirement that only corporate approved callers may submit sensitive requests.
Question 365
Consider University research network and its GKE deployment safety workload. The issue is that a new GPU workload is rolled onto the only node while existing experiment pods have no disruption budget; the current observation is that pilot update evicts every running inference pod. The team must ensure that research service must remain available during cluster upgrades. The change committee needs to understand the current system rather than install a fix yet. What should it do?
- Roll out node-pool updates incrementally with fallback capacity
- Design safe rollout using health probes and PodDisruptionBudget where appropriate
- Set replicas to one and skip readiness probes to accelerate upgrades
- Inventory replicas, node affinity and rescheduling constraints
Explanation
Correct Answer
D:
Inventory replicas, node affinity and rescheduling constraints is the appropriate diagnostic because the observation—pilot update evicts every running inference pod—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (a new GPU workload is rolled onto the only node while existing experiment pods have no disruption budget) before the design is approved.
Incorrect Answers
A:
The alternative “Roll out node-pool updates incrementally with fallback capacity” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, pilot update evicts every running inference pod. The option “Inventory replicas, node affinity and rescheduling constraints” instead addresses the requested root-cause evidence.
B:
The alternative “Design safe rollout using health probes and PodDisruptionBudget where appropriate” is target architecture selection; it does not establish why the observed path failed. In the reported observation, pilot update evicts every running inference pod. The option “Inventory replicas, node affinity and rescheduling constraints” instead addresses the requested root-cause evidence.
C:
Unprotected singleton workloads are unavailable on node drain. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that research service must remain available during cluster upgrades.
Question 366
Retail airline reservations plans to update load-test realism. Investigators found that peak production incident shows locks on seat inventory despite low GET latency, consistent with the reported problem that capacity tests generate simple GET requests but real transactions require multiple write steps. The response must satisfy the requirement that holiday purchase flow must sustain transactional volume without overselling. The approved change has been applied. Which validation provides the clearest evidence the original problem is fixed?
- Run high-concurrency seat purchases and verify one valid reservation per seat
- Introduce production-like synthetic traffic safely before peak season
- Model full booking transactions, contention and peak user behavior
- Declare peak readiness because static homepage GET requests are fast
Explanation
Correct Answer
A:
Run high-concurrency seat purchases and verify one valid reservation per seat is a direct acceptance test of the path implicated by peak production incident shows locks on seat inventory despite low GET latency. It checks that the original risk (capacity tests generate simple GET requests but real transactions require multiple write steps) has actually been controlled and tests that permitted operations still meet the requirement that holiday purchase flow must sustain transactional volume without overselling.
Incorrect Answers
B:
The alternative “Introduce production-like synthetic traffic safely before peak season” documents change execution but not its operational acceptance result. In the reported observation, peak production incident shows locks on seat inventory despite low GET latency. The option “Run high-concurrency seat purchases and verify one valid reservation per seat” instead addresses the requested acceptance verification.
C:
The alternative “Model full booking transactions, contention and peak user behavior” collects prechange evidence without confirming that the revised service meets its outcome. In the reported observation, peak production incident shows locks on seat inventory despite low GET latency. The option “Run high-concurrency seat purchases and verify one valid reservation per seat” instead addresses the requested acceptance verification.
D:
Read-only performance does not establish transactional capacity. For this question, it is not a defensible choice for acceptance verification and it does not satisfy the requirement that holiday purchase flow must sustain transactional volume without overselling.
Question 367
National rail passenger service plans to update disaster recovery runbook. Investigators found that DR rehearsal cannot identify who owns recovery steps, consistent with the reported problem that backup jobs report success but operational teams have never restored ticket data. The response must satisfy the requirement that tickets must be sold again within 30 minutes after a primary failure. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Inventory recovery dependencies, owners, RTO/RPO and validated backups
- Increase the backup frequency and skip every restoration exercise
- Build owner-assigned DR procedures with restoration priorities and decision gates
- Monitor drill age, ownership gaps and recovery time trends
- Schedule isolated recovery drills and sign off runbook changes
Explanation
Correct Answers
C:
Build owner-assigned DR procedures with restoration priorities and decision gates targets the underlying issue (backup jobs report success but operational teams have never restored ticket data) rather than merely reporting symptoms. It also preserves the required outcome (tickets must be sold again within 30 minutes after a primary failure), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
E:
Schedule isolated recovery drills and sign off runbook changes is a controlled means to introduce the approved architecture while meeting the requirement that tickets must be sold again within 30 minutes after a primary failure. 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 (backup jobs report success but operational teams have never restored ticket data), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
B:
Backup completion does not prove recoverability or an executable procedure. The shortcut does not supply the requested design and safe rollout for disaster recovery runbook, where DR rehearsal cannot identify who owns recovery steps.
D:
Monitoring disaster recovery runbook for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: backup jobs report success but operational teams have never restored ticket data.
Question 368
Case study: Medical imaging supplier: patient identifier transformation
Medical imaging supplier relies on patient identifier transformation to deliver its operational commitments. The current architectural problem is that the data warehouse stores masked fields but shared exports retain direct identifiers. The team has recorded this evidence: sampling shows patient names survive in analytical free-text notes. Any proposed response must preserve the requirement that only approved care providers may see reidentifiable data.
Medical imaging supplier plans to update patient identifier transformation. Investigators found that sampling shows patient names survive in analytical free-text notes, consistent with the reported problem that the data warehouse stores masked fields but shared exports retain direct identifiers. The response must satisfy the requirement that only approved care providers may see reidentifiable data. The design review is over and production introduction is next. Which action is the best fit?
- Use Sensitive Data Protection and governed deidentification with appropriate keys
- Stage transformations and validate clinical utility before enabling broad analytics
- Simply rename a column from patient_name to alias
- Track newly identified PII patterns and masking policy exceptions
Explanation
Correct Answer
B:
Stage transformations and validate clinical utility before enabling broad analytics is a controlled means to introduce the approved architecture while meeting the requirement that only approved care providers may see reidentifiable data. Staged delivery, attributable change evidence and rollback guard against an otherwise avoidable service disruption.
Incorrect Answers
A:
The alternative “Use Sensitive Data Protection and governed deidentification with appropriate keys” describes the chosen design, but not how it is introduced safely. In the reported observation, sampling shows patient names survive in analytical free-text notes. The option “Stage transformations and validate clinical utility before enabling broad analytics” instead addresses the requested controlled implementation.
C:
Renaming does not remove personal identifiers. For this question, it is not a defensible choice for controlled implementation and it does not satisfy the requirement that only approved care providers may see reidentifiable data.
D:
The alternative “Track newly identified PII patterns and masking policy exceptions” addresses later monitoring rather than the actual controlled deployment. In the reported observation, sampling shows patient names survive in analytical free-text notes. The option “Stage transformations and validate clinical utility before enabling broad analytics” instead addresses the requested controlled implementation.
Question 369
Case study: Coastal logistics firm: Storage API pagination
Coastal logistics firm relies on Storage API pagination to deliver its operational commitments. The current architectural problem is that inventory script processes only the first page of Cloud Storage objects. The team has recorded this evidence: logs show listing stops at 1,000 objects despite millions in bucket. Any proposed response must preserve the requirement that compliance inventory must enumerate all object generations.
Consider Coastal logistics firm and its Storage API pagination workload. The issue is that inventory script processes only the first page of Cloud Storage objects; the current observation is that logs show listing stops at 1,000 objects despite millions in bucket. The team must ensure that compliance inventory must enumerate all object generations. Select TWO complementary actions: one durable target architecture and one controlled way to introduce it while preserving service. Which TWO are best?
- Assume one list call always returns an entire bucket
- Watch page-token errors and missed generations
- Use the supported client-library iterator with explicit generation handling
- Inspect pagination tokens, versions and API quota behavior
- Deploy incremental inventory with restartable page checkpoints
Explanation
Correct Answers
C:
Use the supported client-library iterator with explicit generation handling targets the underlying issue (inventory script processes only the first page of Cloud Storage objects) rather than merely reporting symptoms. It also preserves the required outcome (compliance inventory must enumerate all object generations), which a target-state design must satisfy. This supplies the lasting-control half of the two-action package.
E:
Deploy incremental inventory with restartable page checkpoints is a controlled means to introduce the approved architecture while meeting the requirement that compliance inventory must enumerate all object generations. 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:
Paginated APIs can omit most objects unless continued. The shortcut does not supply the requested design and safe rollout for Storage API pagination, where logs show listing stops at 1,000 objects despite millions in bucket.
B:
Monitoring Storage API pagination for future drift is useful after the rollout, but it does not specify the requested target control and safe implementation for this case: inventory script processes only the first page of Cloud Storage objects.
D:
This diagnoses the original issue (inventory script processes only the first page of Cloud Storage objects), but the TWO requested actions are the target architecture and its implementation; discovery was already completed.
Question 370
Case study: National pharmacy network: central NAT egress
National pharmacy network relies on central NAT egress to deliver its operational commitments. The current architectural problem is that a vendor allowlist changes every time cloud VMs autoscale with ephemeral external IPs. The team has recorded this evidence: packet trace shows outbound API requests use different VM public addresses. Any proposed response must preserve the requirement that drug inventory partner permits only named stable source addresses.
At National pharmacy network, the Google Cloud design for central NAT egress is affected because a vendor allowlist changes every time cloud VMs autoscale with ephemeral external IPs. A review found that packet trace shows outbound API requests use different VM public addresses. The proposed response must account for the requirement that drug inventory partner permits only named stable source addresses. Which analysis will allow the architect to distinguish the reported symptom from the actual failure path?
- Assign a new public IP to every ephemeral VM and email the vendor daily
- Migrate VMs to private instances and update partner allowlists before cutover
- Identify egress flows and where dynamic external addressing is introduced
- Use Cloud NAT with reserved egress IP addresses for private-subnet instances
Explanation
Correct Answer
C:
Identify egress flows and where dynamic external addressing is introduced is the appropriate diagnostic because the observation—packet trace shows outbound API requests use different VM public addresses—does not by itself establish every dependency, authorization rule or failure boundary. The findings explain the reported risk (a vendor allowlist changes every time cloud VMs autoscale with ephemeral external IPs) before the design is approved.
Incorrect Answers
A:
Changing per-instance IPs cannot maintain a stable partner allowlist. For this question, it is not a defensible choice for root-cause evidence and it does not satisfy the requirement that drug inventory partner permits only named stable source addresses.
B:
The alternative “Migrate VMs to private instances and update partner allowlists before cutover” is rollout work that may modify the wrong dependency before root-cause evidence is collected. In the reported observation, packet trace shows outbound API requests use different VM public addresses. The option “Identify egress flows and where dynamic external addressing is introduced” instead addresses the requested root-cause evidence.
D:
The alternative “Use Cloud NAT with reserved egress IP addresses for private-subnet instances” is target architecture selection; it does not establish why the observed path failed. In the reported observation, packet trace shows outbound API requests use different VM public addresses. The option “Identify egress flows and where dynamic external addressing is introduced” instead addresses the requested root-cause evidence.