PCDBE practice questions 21–30

1. Question 21 · 170 in bank

   You are configuring the networking of a Cloud SQL instance. The only application that connects to this database resides on a Compute Engine VM in the same project as the Cloud SQL instance. The VM and the Cloud SQL instance both use the same VPC network, and both have an external (public) IP address and an internal (private) IP address. You want to improve network security. What should you do?

   • ADisable and remove the internal IP address assignment.
   • BDisable both the external IP address and the internal IP address, and instead rely on Private Google Access.
   • CSpecify an authorized network with the CIDR range of the VM.
   • DDisable and remove the external IP address assignment.

   Show answer

   Answer: D

   Explanation

   D. It is always more secure to use an internal IP, so removing them doesn’t make sense. Eliminate A. You can use Private Google Access when VM instances only have internal IP addresses, so disabling the internal IPs and use Private Google Access doesn’t make sense. Eliminate B. Specifying an authorized network when they’re on the same subnet doesn’t make sense. Eliminate C. A way to improve network security would be to disable external IPs since they’re not needed.

   Source: Adapted from community discussion — dynamic_dba

   Next question

2. Question 22 · 170 in bank

   You are managing two different applications: Order Management and Sales Reporting. Both applications interact with the same Cloud SQL for MySQL database. The Order Management application reads and writes to the database 24/7, but the Sales Reporting application is read-only. Both applications need the latest data. You need to ensure that the Performance of the Order Management application is not affected by the Sales Reporting application. What should you do?

   • ACreate a read replica for the Sales Reporting application.
   • BCreate two separate databases in the instance, and perform dual writes from the Order Management application.
   • CUse a Cloud SQL federated query for the Sales Reporting application.
   • DQueue up all the requested reports in PubSub, and execute the reports at night.

   Show answer

   Answer: A

   Explanation

   A. B could be done, but would need some application work to make it happen. Eliminate B. Federated queries are queries sent from BigQuery to Cloud SQL or Cloud Spanner with the results returned to BigQuery as a temporary table. So C is wrong. Even if D made sense, running all the reports at night would still impact the Order Management application since it accesses the database 24x7. That leaves A which is the classic solution to offload reads from the primary instance.

   Source: Adapted from community discussion — dynamic_dba

   Next question

3. Question 23 · 170 in bank

   You are the DBA of an online tutoring application that runs on a Cloud SQL for PostgreSQL database. You are testing the implementation of the cross-regional failover configuration. The database in region R1 fails over successfully to region R2, and the database becomes available for the application to process data. During testing, certain scenarios of the application work as expected in region R2, but a few scenarios fail with database errors. The application-related database queries, when executed in isolation from Cloud SQL for PostgreSQL in region R2, work as expected. The application performs completely as expected when the database fails back to region R1. You need to identify the cause of the database errors in region R2. What should you do?

   • ADetermine whether the versions of Cloud SQL for PostgreSQL in regions R1 and R2 are different.
   • BDetermine whether the database patches of Cloud SQI for PostgreSQL in regions R1 and R2 are different.
   • CDetermine whether the failover of Cloud SQL for PostgreSQL from region R1 to region R2 is in progress or has completed successfully.
   • DDetermine whether Cloud SQL for PostgreSQL in region R2 is a near-real-time copy of region R1 but not an exact copy.

   Show answer

   Answer: D

   Explanation

   Verify that the replica has processed all the transactions it has received from the primary. This ensures that when promoted, the replica reflects all transactions that were received before the primary became unavailable. https://cloud.google.com/sql/docs/postgres/replication/cross-region-replicas#verify_failover_criteria

   Source: Adapted from community discussion — chelbsik

   References

   • Google Cloud documentation
   • Google Cloud documentation

   Next question

4. Question 24 · 170 in bank

   You are designing an augmented reality game for iOS and Android devices. You plan to use Cloud Spanner as the primary backend database for game state storage and player authentication. You want to track in-game rewards that players unlock at every stage of the game. During the testing phase, you discovered that costs are much higher than anticipated, but the query response times are within the SLA. You want to follow Google-recommended practices. You need the database to be performant and highly available while you keep costs low. What should you do?

   • AManually scale down the number of nodes after the peak period has passed.
   • BUse interleaving to co-locate parent and child rows.
   • CUse the Cloud Spanner query optimizer to determine the most efficient way to execute the SQL query.
   • DUse granular instance sizing in Cloud Spanner and Autoscaler.

   Show answer

   Answer: D

   Explanation

   D. A is nonsense. Using interleaved tables can help speed up queries, but the question says query response times are OK. So B is wrong. C is wrong for the same reason. That leaves D. The question is about which factors determine the cost of running Spanner. They include region vs. multi-region, compute unit (nodes or processing units), how much storage and how much backup space. From the Google docs, it says “When you create a Cloud Spanner instance, you choose the number of compute capacity nodes or processing units to serve your data. However, if the workload of an instance changes, Cloud Spanner doesn't automatically adjust the size of the instance. This document introduces the Autoscaler tool for Cloud Spanner (Autoscaler), an open source tool that you can use as a companion tool to Cloud Spanner. This tool lets you automatically increase or reduce the number of nodes or processing units in one or more Spanner instances based on how their capacity is being used.” https://cloud.google.com/spanner/docs/autoscaling-overview

   Source: Adapted from community discussion — dynamic_dba

   References

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   Next question

5. Question 25 · 170 in bank

   Your company wants to migrate its MySQL, PostgreSQL, and Microsoft SQL Server on-premises databases to Google Cloud. You need a solution that provides near-zero downtime, requires no application changes, and supports change data capture (CDC). What should you do?

   • AUse the native export and import functionality of the source database.
   • BCreate a database on Google Cloud, and use database links to perform the migration.
   • CCreate a database on Google Cloud, and use Dataflow for database migration.
   • DUse Database Migration Service.

   Show answer

   Answer: D

   Next question

6. Question 26 · 170 in bank

   You are managing a Cloud SQL for PostgreSQL instance in Google Cloud. You have a primary instance in region 1 and a read replica in region 2. After a failure of region 1, you need to make the Cloud SQL instance available again. You want to minimize data loss and follow Google-recommended practices. What should you do?

   • ARestore the Cloud SQL instance from the automatic backups in region 3.
   • BRestore the Cloud SQL instance from the automatic backups in another zone in region 1.
   • CCheck "Lag Bytes" in the monitoring dashboard for the primary instance in the read replica instance. Check the replication status using pg_catalog.pg_last_wal_receive_lsn(). Then, fail over to region 2 by promoting the read replica instance.
   • DCheck your instance operational log for the automatic failover status. Look for time, type, and status of the operations. If the failover operation is successful, no action is necessary. Otherwise, manually perform gcloud sql instances failover .

   Show answer

   Answer: C

   Explanation

   C. No need for restores if you have a read replica which you can promote to be the new primary. Eliminate A and B. Failovers do not happen automatically to read replicas. You have to promote them. Eliminate D. That leaves C which is supported by Google's documentation. https://cloud.google.com/sql/docs/postgres/replication/cross-region-replicas#disaster_recovery

   Source: Adapted from community discussion — dynamic_dba

   References

   • Google Cloud documentation

   Next question

7. Question 27 · 170 in bank

   You need to issue a new server certificate because your old one is expiring. You need to avoid a restart of your Cloud SQL for MySQL instance. What should you do in your Cloud SQL instance?

   • AIssue a rollback, and download your server certificate.
   • BCreate a new client certificate, and download it.
   • CCreate a new server certificate, and download it.
   • DReset your SSL configuration, and download your server certificate.

   Show answer

   Answer: C

   Explanation

   C. Issuing a rollback is nonsense. Eliminate A. The client certificate is unaffected. Eliminate B. Resetting the SSL configuration is not mentioned, but if you enforced SSL that would require an instance re-start. All you need to do is create a new server certificate and download it (to your clients). https://cloud.google.com/sql/docs/sqlserver/configure-ssl-instance#server-certs

   Source: Adapted from community discussion — dynamic_dba

   References

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   Next question

8. Question 28 · 170 in bank

   Your company is migrating all legacy applications to Google Cloud. All on-premises applications are using legacy Oracle 12c databases with Oracle Real Application Cluster (RAC) for high availability (HA) and Oracle Data Guard for disaster recovery. You need a solution that requires minimal code changes, provides the same high availability you have today on-premises, and supports a low latency network for migrated legacy applications. What should you do?

   • AMigrate the databases to Cloud Spanner.
   • BMigrate the databases to Cloud SQL, and enable a standby database.
   • CMigrate the databases to Compute Engine using regional persistent disks.
   • DMigrate the databases to Bare Metal Solution for Oracle.

   Show answer

   Answer: D

   Explanation

   D. BMS is the only Google database service which supports Oracle aside from GCVE. It allows you to use all native Oracle features including RAC. Since GCVE isn't mentioned, it has to be D - Bare Metal Solution.

   Source: Adapted from community discussion — dynamic_dba

   Next question

9. Question 29 · 170 in bank

   Your company is evaluating Google Cloud database options for a mission-critical global payments gateway application. The application must be available 24/7 to users worldwide, horizontally scalable, and support open source databases. You need to select an automatically shardable, fully managed database with 99.999% availability and strong transactional consistency. What should you do?

   • ASelect Bare Metal Solution for Oracle.
   • BSelect Cloud SQL.
   • CSelect Bigtable.
   • DSelect Cloud Spanner.

   Show answer

   Answer: D

   Explanation

   D. Oracle is not open source. Eliminate A. Horizontally scalable means adding more servers to run the database. That eliminates Cloud SQL (B). Strong transactional consistency rules out Bigtable since Bigtable transactions are at the single row level. Eliminate C. That leaves D.

   Source: Adapted from community discussion — dynamic_dba

   References

   • Google Cloud documentation

   Next question

10. Question 30 · 170 in bank

    You are a DBA of Cloud SQL for PostgreSQL. You want the applications to have password-less authentication for read and write access to the database. Which authentication mechanism should you use?

    • AUse Identity and Access Management (IAM) authentication.
    • BUse Managed Active Directory authentication.
    • CUse Cloud SQL federated queries.
    • DUse PostgreSQL database's built-in authentication.

    Show answer

    Answer: A

    Explanation

    A. Google wants you to use IAM regardless. PostgreSQL built-in authentication requires a username and a password. That rules out D. Federated queries are queries originating from BigQuery to Cloud SQL. Eliminate C. AD is strongly tied to SQL Server. Eliminate B. That leaves A.

    Source: Adapted from community discussion — dynamic_dba

    References

    • Google Cloud documentation

    Next question

Keyboard:J Kmove between questionsA–Dselect an answerRshow or hide the answer← →change page

PCDBE practice, question 21. 170 questions available.