Question 470
A transactional service must fail over automatically when a zone hosting its primary database becomes unavailable. How should the team proceed?
- Enable supported regional HA for the database service
- Use a supported Cloud SQL read replica for read traffic
- Design supported cross-region replication and backup schedules to match the RPO
- Use a valid point-in-time backup or recovery mechanism
Explanation
Correct Answer
A:Regional HA distributes critical database components across zones for certain failure scenarios. It does not provide a blanket guarantee against accidental deletes.
Incorrect Answers
B:This mechanism is intended to scale relational database reads. It does not implement the required action to maintain service through a zonal outage, even though both are relevant to the same general service.
C:The tool can meet a regional database recovery point target, but it cannot be substituted for the capability needed to maintain service through a zonal outage. Its operational effect is not equivalent.
D:Select this when the goal is to recover pre-corruption database state. Here the decisive requirement is to maintain service through a zonal outage, and that outcome is not achieved by this action.
Question 471
A team requires database recovery from logical corruption that was replicated to standbys. What is the most direct solution?
- Use a valid point-in-time backup or recovery mechanism
- Enable supported regional HA for the database service
- Connect with an authorized SQL client or Cloud SQL connector
- Design supported cross-region replication and backup schedules to match the RPO
- Use a supported Cloud SQL read replica for read traffic
Explanation
Correct Answer
A:HA replicas can copy bad writes along with good ones. Backups and PITR address logical recovery points rather than only infrastructure availability.
Incorrect Answers
B:This approach can provide automated database zone failover. The requirement is to undo replicated data loss, so it controls a different part of the workflow and leaves that result unaddressed.
C:An operator would use this action to execute SQL against Cloud SQL, chiefly when trying to query an operational relational database; The current requirement is to undo replicated data loss, which remains unresolved by that operation.
D:This setting would meet a regional database recovery point target. That behavior does not establish the requested ability to undo replicated data loss; it addresses an adjacent concern instead.
E:Its purpose is to scale relational database reads. That distinction matters: the scenario calls for an action to undo replicated data loss, not this separate configuration task.
Question 472
A database must meet an explicit recovery point objective across regions. Which implementation matches this requirement?
- Design supported cross-region replication and backup schedules to match the RPO
- Enable supported regional HA for the database service
- Use a supported Bigtable client to read the keyed row
- Use a supported Cloud SQL read replica for read traffic
- Use a valid point-in-time backup or recovery mechanism
Explanation
Correct Answer
A:Region redundancy depends on product features and replication lag. A single-region HA configuration alone does not define remote recovery capability.
Incorrect Answers
B:This mechanism is intended to provide automated database zone failover. It does not implement the required action to plan cross-region disaster recovery, even though both are relevant to the same general service.
C:For the stated need to plan cross-region disaster recovery, this would instead read a Bigtable row by key; it supports the separate task to retrieve wide-column data efficiently, not the required configuration.
D:Select this when the goal is to scale relational database reads. Here the decisive requirement is to plan cross-region disaster recovery, and that outcome is not achieved by this action.
E:Administrators use it to recover pre-corruption database state, rather than to plan cross-region disaster recovery. Applying it would change the wrong component of this deployment.
Question 473
A streaming pipeline appears stuck and the operator needs the current Dataflow job state. Which action is most appropriate?
- Inspect the BigQuery load job error details and adjust the schema or source
- View the BigQuery job errors and principal context
- Inspect pipeline metrics, dead-letter handling, and destination write errors
- Inspect the Dataflow job details and worker logs
Explanation
Correct Answer
D:Dataflow job monitoring exposes job state and step-level information. It is distinct from the source Pub/Sub topic's metadata.
Incorrect Answers
A:It can troubleshoot BigQuery load jobs; however, it does not provide the specific operation to diagnose a stalled Dataflow job. The two controls serve different purposes.
B:This choice is useful to inspect BigQuery access failures. The stated task is to diagnose a stalled Dataflow job, which requires a different mechanism or permission scope.
C:This setting would audit Dataflow output processing. That behavior does not establish the requested ability to diagnose a stalled Dataflow job; it addresses an adjacent concern instead.
Question 474
An analytics query failed with an access error and the team needs to confirm its execution metadata. Which solution should be used?
- Inspect pipeline metrics, dead-letter handling, and destination write errors
- Inspect the Dataflow job details and worker logs
- Grant a suitable Storage Object Viewer role on the specific bucket
- Inspect the BigQuery load job error details and adjust the schema or source
- View the BigQuery job errors and principal context
Explanation
Correct Answer
E:BigQuery job metadata records error details. Permissions to submit a job and read referenced datasets can be independent.
Incorrect Answers
A:Administrators use it to audit Dataflow output processing, rather than to investigate a denied query job. Applying it would change the wrong component of this deployment.
B:While this can inspect managed pipeline execution, the present task requires an action to investigate a denied query job. This operation does not supply the behavior described in the requirements.
C:The result of this choice would be to grant read access to objects in one bucket. The requested outcome, to investigate a denied query job, is handled at a different layer or through another operation.
D:This mechanism is intended to troubleshoot BigQuery load jobs. It does not implement the required action to investigate a denied query job, even though both are relevant to the same general service.
Question 475
A batch transform finished but the output table has fewer rows than expected. How should the team proceed?
- Inspect the Dataflow job details and worker logs
- Inspect pipeline metrics, dead-letter handling, and destination write errors
- Estimate query bytes processed and apply partitioning or clustering where suitable
- View the BigQuery job errors and principal context
- Inspect the BigQuery load job error details and adjust the schema or source
Explanation
Correct Answer
B:Dataflow pipeline metrics and logs help locate discarded or failed records. A completed job state does not by itself guarantee the expected business-level row count.
Incorrect Answers
A:This choice is useful to inspect managed pipeline execution. The stated task is to explain unexpected ETL row counts, which requires a different mechanism or permission scope.
C:For the stated need to explain unexpected ETL row counts, this would instead optimize warehouse query scanning cost; it supports the separate task to lower BigQuery query consumption, not the required configuration.
D:It can inspect BigQuery access failures; however, it does not provide the specific operation to explain unexpected ETL row counts. The two controls serve different purposes.
E:Select this when the goal is to troubleshoot BigQuery load jobs. Here the decisive requirement is to explain unexpected ETL row counts, and that outcome is not achieved by this action.
Question 476
A warehouse import has a schema mismatch and failed during load. What is the most direct solution?
- Inspect the BigQuery load job error details and adjust the schema or source
- Analyze storage-class retrieval and operations costs before changing its class
- View the BigQuery job errors and principal context
- Inspect the Dataflow job details and worker logs
- Inspect pipeline metrics, dead-letter handling, and destination write errors
Explanation
Correct Answer
A:Load jobs validate source formats and target schemas. Changing a VM firewall rule cannot repair a mismatched CSV field type.
Incorrect Answers
B:In a workflow for storage cost analysis, it can assess retrieval cost in cold storage; This question instead asks how to fix ingestion schema errors; substituting the former action would leave that distinct technical requirement unmet.
C:While this can inspect BigQuery access failures, the present task requires an action to fix ingestion schema errors. This operation does not supply the behavior described in the requirements.
D:This approach can inspect managed pipeline execution. The requirement is to fix ingestion schema errors, so it controls a different part of the workflow and leaves that result unaddressed.
E:This setting would audit Dataflow output processing. That behavior does not establish the requested ability to fix ingestion schema errors; it addresses an adjacent concern instead.
Question 477
An archive workload retains objects for long periods and rarely reads them. Finance is deciding whether to move those objects from Standard to a colder Cloud Storage class. Which TWO cost dimensions must it evaluate? (Choose TWO.)
- The number of IAM basic Owner roles in the project as the sole storage-class price factor
- Expected retrieval and operation charges when objects are read or manipulated
- Whether a Cloud DNS public zone is configured for the bucket's application hostname
- Minimum storage duration and any applicable early-deletion charges
Explanation
Correct Answers
B:Colder storage classes can lower at-rest charges while increasing certain access charges. The workload's actual read and operation patterns should be included in total-cost estimates.
D:Nearline, Coldline and Archive have minimum storage duration rules. Deleting or rewriting objects early can affect the financial benefit of the class change.
Incorrect Answers
A:IAM role counts are not the primary mechanism for calculating Cloud Storage storage-class costs. Data volume, access, operations, duration and location matter.
C:Cloud DNS concerns name resolution. It does not eliminate storage-class minimum-duration requirements or Cloud Storage retrieval charges.
Question 478
A dataset has many tiny objects and costs are dominated by frequent storage operations. Which action is most appropriate?
- Analyze storage-class retrieval and operations costs before changing its class
- Include applicable network egress and retrieval charges in the estimate
- Review Cloud Storage operations pricing and request patterns
- Estimate query bytes processed and apply partitioning or clustering where suitable
Explanation
Correct Answer
C:Object operation charges can matter in addition to byte-month storage. Changing a region without addressing request volume may not solve the cause.
Incorrect Answers
A:Using this action would assess retrieval cost in cold storage, not identify API operation driven expenses. Its function is different from the specific change described in the scenario.
B:This choice is useful to estimate object transfer and retrieval cost. The stated task is to identify API operation driven expenses, which requires a different mechanism or permission scope.
D:This setting would optimize warehouse query scanning cost. That behavior does not establish the requested ability to identify API operation driven expenses; it addresses an adjacent concern instead.
Question 479
A team wants to estimate how cross-region object downloads will affect expenses. Which solution should be used?
- Estimate query bytes processed and apply partitioning or clustering where suitable
- Enable and review appropriate Cloud Audit Logs for Cloud Storage access
- Analyze storage-class retrieval and operations costs before changing its class
- Review Cloud Storage operations pricing and request patterns
- Include applicable network egress and retrieval charges in the estimate
Explanation
Correct Answer
E:Data transfer routes and storage class affect billed reads. Stored capacity cost by itself is an incomplete model for downloads.
Incorrect Answers
A:This mechanism is intended to optimize warehouse query scanning cost. It does not implement the required action to price cross-region data access, even though both are relevant to the same general service.
B:In a workflow for object retention and protection distinctions, it can capture object access audit records; This question instead asks how to price cross-region data access; substituting the former action would leave that distinct technical requirement unmet.
C:While this can assess retrieval cost in cold storage, the present task requires an action to price cross-region data access. This operation does not supply the behavior described in the requirements.
D:Administrators use it to analyze object request cost, rather than to price cross-region data access. Applying it would change the wrong component of this deployment.
Question 480
A BigQuery cost review must choose between scanning less data and changing compute infrastructure. How should the team proceed?
- Analyze storage-class retrieval and operations costs before changing its class
- Estimate query bytes processed and apply partitioning or clustering where suitable
- Review Cloud KMS key-version disablement effects before changing it
- Include applicable network egress and retrieval charges in the estimate
- Review Cloud Storage operations pricing and request patterns
Explanation
Correct Answer
B:BigQuery query billing and performance are influenced by the workload and pricing model. Better query pruning can reduce scanned data without provisioning VMs.
Incorrect Answers
A:Using this action would assess retrieval cost in cold storage, not lower BigQuery query consumption. Its function is different from the specific change described in the scenario.
C:The choice is aimed at a different purpose: to control access through key lifecycle; Even if it can manage CMEK key version availability, that would not accomplish the stated requirement to lower BigQuery query consumption.
D:It can estimate object transfer and retrieval cost; however, it does not provide the specific operation to lower BigQuery query consumption. The two controls serve different purposes.
E:This choice is useful to analyze object request cost. The stated task is to lower BigQuery query consumption, which requires a different mechanism or permission scope.