PCDE Practice Test 47 — 15 Questions

Question 1

A retail company runs a Cloud SQL for MySQL instance that supports a customer-facing ordering system. The database team wants to improve availability and reduce downtime during maintenance and zonal failures. They are evaluating High Availability configuration options. Which TWO statements accurately describe Cloud SQL for MySQL High Availability? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

A standby instance is created in a different zone within the same region, and it is kept in sync with the primary.. Cloud SQL for MySQL High Availability creates a synchronized standby in a different zone of the same region and promotes it automatically on failure while preserving the instance connection name. The standby is not a read-scaling resource, backups are still required for point-in-time recovery, and toggling HA on an existing instance causes a restart, so only the zone-redundancy and automatic-failover statements are accurate.

Answer

The standby instance can be used to serve read-only queries to reduce load on the primary.

Answer

High Availability can be enabled or disabled on an existing instance without causing any interruption to connections.

Answer

Enabling High Availability eliminates the need for backups because the standby holds a complete copy of the data.

Question 2

A healthcare startup runs an AlloyDB for PostgreSQL cluster and wants to offload read-only analytics queries from the primary instance. The queries are long-running and tolerate some replication lag. What should they do?

Accepted Answer

Create a read pool instance and direct analytics queries to its read endpoint.. AlloyDB read pool instances replicate from the primary and serve read-only queries through dedicated endpoints. Directing long-running analytics to a read pool isolates that workload from transactional traffic on the primary and scales reads horizontally. Because the queries tolerate replication lag, the asynchronous nature of read pool replication is acceptable here.

Answer

Add additional vCPUs to the primary instance and enable columnar engine on the primary.

Answer

Enable a cross-region secondary cluster and route analytics queries to its primary instance.

Answer

Configure a Cloud SQL read replica of the AlloyDB primary and point analytics queries at it.

Question 3

An engineer needs to migrate a MySQL database to Cloud SQL with minimal downtime. Which TWO steps should be part of the migration plan? (Choose 2)

Accepted Answer

Verify that the source MySQL version is compatible with Cloud SQL. Option C is correct because Cloud SQL supports only specific MySQL major/minor versions, so verifying that the source MySQL version is compatible with the target Cloud SQL version is a mandatory prerequisite before any migration can succeed. Option D is correct because Database Migration Service (DMS) provides continuous replication from the source MySQL to Cloud SQL, allowing a cutover with minimal downtime rather than a one-time dump-and-load. Option A is not appropriate because a Cloud SQL read replica can only be created from another Cloud SQL instance, not directly from an external/on-premises MySQL source. Option B is not suitable for minimal downtime because mysqldump plus import requires taking the source offline or accepting data loss during the dump window. Option E is unnecessary because DMS handles connectivity (for example via IP allowlisting or a reverse SSH tunnel) and a Cloud VPN tunnel is not a required migration step.

Answer

Create a Cloud SQL read replica from the source

Answer

Perform a mysqldump and import the dump into Cloud SQL

Answer

Set up a Cloud VPN tunnel between on-premise and GCP

Question 4

Your Cloud Bigtable instance uses HDD storage. You need to change to SSD to improve read performance. What is the correct procedure?

Accepted Answer

Create a new cluster with SSD storage in the same instance, replicate data, then delete the old cluster.. Cloud Bigtable does not support changing the storage type of an existing cluster in place. The supported migration path is to add a new cluster with SSD storage to the same instance, allow Bigtable's built-in replication to sync the data, then delete the old HDD cluster. This preserves data and avoids downtime because the instance continues serving reads/writes throughout the replication and cutover.

Answer

Use Cloud Bigtable's online storage migration feature to switch to SSD without downtime.

Answer

Update the instance's storage type using 'gcloud bigtable instances update' with the --storage-type flag.

Answer

Export the HDD cluster's data to Cloud Storage, then import into a new SSD cluster.

Question 5

A media company runs a Cloud SQL for PostgreSQL instance in us-central1 that serves a read-heavy reporting dashboard. The dashboard queries have grown to consume most of the CPU on the primary instance, degrading write performance for the transactional application. The team wants to offload reporting reads without changing the application's write connection string, and wants failover protection if the primary zone fails. Which configuration should they implement?

Accepted Answer

Enable high availability on the primary instance and create a same-region read replica, then point the reporting dashboard's read-only connection string to the read replica.. A same-region read replica offloads read-heavy reporting traffic with low replication lag, while enabling high availability on the primary provides an automatic failover standby in a different zone. Together they satisfy both the read-scaling and zone-resilience requirements without changing the transactional application's write connection string.

Answer

Increase the primary instance's vCPU count and memory, then enable point-in-time recovery so the dashboard reads from the primary without contention.

Answer

Create a cross-region read replica in us-east1 and point the reporting dashboard's read-only connection string to that replica's IP address.

Answer

Convert the primary instance to a read replica of a new primary in another region, and have the dashboard connect to the new primary.

Question 6

A retail company runs its order-processing service on a Cloud SQL for PostgreSQL Enterprise Plus instance in us-east1. The application team reports that during a recent zone-level outage the instance failed over, but the failover took several minutes and the connection string had to be changed manually. The company wants the database to fail over automatically to another zone without any application-side connection changes and to minimize the failover time. What should the database engineer configure?

Accepted Answer

Enable high availability on the instance by configuring a regional instance with a primary and standby zone.. Automatic, application-transparent failover in Cloud SQL is delivered by a regional high-availability configuration, which maintains a standby in a separate zone behind the same IP address. Because the IP does not change during promotion, the application reconnects without editing connection strings. Read replicas, larger single-zone instances, and connection proxies do not provide the required automatic promotion behavior.

Answer

Configure a Cloud SQL Auth Proxy sidecar in the application and point it at the instance connection name.

Answer

Move the instance to a single zone with a larger machine type and enable automatic storage increase.

Answer

Create a read replica in a second zone and update the application to write to the replica when the primary is unavailable.

Question 7

A retail company runs its order-management service on Cloud SQL for PostgreSQL in the europe-west1 region. The database must remain writable during a zonal outage with minimal administrative overhead, and the application connects through a single stable endpoint. Which configuration should the database engineer choose?

Accepted Answer

Configure the instance for high availability (regional) with a primary and standby in two zones of europe-west1, and connect using the instance connection name.. High availability (regional) Cloud SQL for PostgreSQL maintains a synchronous standby in a different zone of the same region and fails over automatically when the primary zone fails. The instance connection name remains valid after failover, so the application keeps one stable endpoint. This satisfies the writable-during-zonal-outage requirement while keeping administrative effort low, unlike zonal instances, read replicas, or self-managed compute.

Answer

Configure the instance as a zonal Cloud SQL for PostgreSQL instance and add a cross-region read replica in europe-west4.

Answer

Deploy the database on a Compute Engine instance with a local SSD and use a managed instance group to restart the VM in another zone.

Answer

Create a Cloud SQL for PostgreSQL instance with a read replica in a second zone of europe-west1 and point the application's write traffic at the read replica.

Question 8

A healthcare analytics company runs Cloud Spanner for a patient records system that must remain available even if an entire region becomes unreachable. The team wants reads to be served from the region closest to each user while writes remain strongly consistent, and they want the configuration to tolerate regional failure without data loss. Which Spanner configuration should the database engineer choose?

Accepted Answer

A multi-region configuration with read-write replicas in multiple regions and read-only replicas placed near user populations.. Cloud Spanner multi-region configurations place read-write replicas across multiple regions to survive a regional outage while preserving strong consistency, and they support adding read-only replicas in other regions for low-latency local reads. A regional configuration cannot survive region loss, and splitting data across separate instances sacrifices Spanner's transactional guarantees.

Answer

A multi-region configuration with only read-only replicas in every region and a single read-write replica.

Answer

A regional configuration with three read-write replicas in three zones of one region.

Answer

Two independent regional instances in different regions with application-level dual writes to both.

Question 9

A gaming company uses Cloud Bigtable for player session data. They have a single cluster in us-central1 and want to add a second cluster in us-east1 for disaster recovery. They need read requests from the us-east1 application tier to be served locally with the lowest possible latency, and they want writes to be replicated automatically. What should they do?

Accepted Answer

Create a second cluster in us-east1 in the same instance and use an app profile with single-cluster routing for the us-east1 application, pointing it to the us-east1 cluster.. Adding a second cluster in us-east1 within the same Bigtable instance enables automatic write replication. An app profile with single-cluster routing pins the us-east1 application's reads to the local cluster, minimizing latency while writes propagate to both clusters. This satisfies both local read latency and automatic replication.

Answer

Create a second cluster in us-east1 and enable a read replica setting on the us-central1 cluster so the us-east1 cluster serves read-only traffic.

Answer

Create a second instance in us-east1 and use the Bigtable replication tool to synchronize tables between the two instances.

Answer

Create a second cluster in us-east1 and configure an app profile with multi-cluster routing for the us-east1 application so reads are distributed across both clusters.

Question 10

A logistics company runs a regional Cloud Spanner instance in us-central1 with processing units sized for current traffic. The team needs the database to survive the loss of an entire zone within the region without any data loss and without manual intervention, and the application must keep writing. Which action should the database engineer take?

Accepted Answer

Keep the regional instance as is, because a regional Cloud Spanner instance already replicates data across three zones within the region.. A regional Cloud Spanner instance already places three read-write replicas across three zones in the region and commits writes through Paxos, so a single-zone failure leaves a quorum that continues serving writes with strong consistency and no data loss. No manual failover, backup restore, or multi-region conversion is needed to survive a zonal outage, which is precisely the resilience a regional configuration provides.

Answer

Add read-only replicas in additional zones and direct all writes to those replicas.

Answer

Create a Cloud Spanner backup schedule and restore the database into a new instance when a zone fails.

Answer

Convert the instance to a multi-region configuration spanning us-central1 and us-east1.

Question 11

A financial services company needs to deploy a Cloud SQL for PostgreSQL instance that must be highly available across two zones within the same region. They want automatic failover with minimal data loss. Which configuration should they choose?

Accepted Answer

A high availability (HA) configuration with a standby instance in a different zone.. For high availability with automatic failover and minimal data loss in Cloud SQL, the correct configuration is a high availability setup with a standby in a different zone. This uses synchronous replication and automatic failover. Other options either lack automatic failover, use asynchronous replication, or are not valid configurations.

Answer

A multi-zone instance with cross-region replication.

Answer

A single-zone instance with automated backups enabled.

Answer

A read replica in a different region.

Question 12

A healthcare company stores sensitive patient data in Cloud Firestore. They need to ensure that data is encrypted with a customer-managed encryption key (CMEK) that they control and can rotate independently. They also need to be able to revoke access to the data by disabling the key. What should they do?

Accepted Answer

Configure Cloud Firestore to use a CMEK from Cloud KMS.. Cloud Firestore supports customer-managed encryption keys (CMEK) through Cloud KMS. Configuring the Firestore database to use a CMEK allows the company to control key rotation and revoke access by disabling the key. Other options either do not provide customer control or add unnecessary complexity.

Answer

Use Cloud External Key Manager (EKM) to store keys outside Google Cloud.

Answer

Encrypt the data at the application level before writing to Firestore.

Answer

Use Google-managed encryption keys and rely on Google's key rotation.

Question 13

A media company uses Bigtable for storing user interaction events. The table has a row key that starts with a timestamp, causing hot spotting and poor performance. They need to redesign the row key to distribute writes evenly while still allowing efficient queries for the most recent events for a given user. What should they do?

Accepted Answer

Use a row key that starts with the user ID, followed by a reversed timestamp.. To avoid hot spotting and allow efficient per-user recent-event queries, the row key should start with the user ID to group a user's events, and then use a reversed timestamp to sort recent events first. This design distributes writes across different users and supports efficient range scans for a user's latest events.

Answer

Reverse the timestamp and prepend it to the row key.

Answer

Use a row key that starts with a hash of the user ID, followed by the timestamp.

Answer

Use a row key that starts with a random number, followed by the user ID and timestamp.

Question 14

A startup is deploying a new Cloud SQL for MySQL instance for a development environment. They want to minimize costs while ensuring that the instance can be automatically patched during a maintenance window. They do not need high availability or read replicas. Which configuration should they choose?

Accepted Answer

A single-zone instance with automated backups enabled and a maintenance window set.. For a development environment with no high availability requirement, a single-zone instance is the most cost-effective. Automated backups provide basic data protection, and setting a maintenance window ensures that automatic patching occurs during a controlled period. Other options either add unnecessary cost or fail to address the maintenance window requirement.

Answer

A single-zone instance with point-in-time recovery enabled and no maintenance window.

Answer

A single-zone instance with binary logging enabled and no maintenance window.

Answer

A regional instance with high availability enabled and a maintenance window set.

Question 15

A healthcare company uses Cloud Bigtable to store patient monitoring data. They need to ensure that the data is replicated across multiple regions for disaster recovery and that reads can be served from the nearest region. They also want to minimize write latency for applications in both the US and Europe. Which Bigtable configuration should they use?

Accepted Answer

A multi-cluster instance with clusters in us-central1 and europe-west1, using a multi-region replication configuration.. Cloud Bigtable multi-cluster instances allow you to create clusters in multiple regions within the same instance. Replication is automatic, providing disaster recovery and enabling reads from the nearest cluster. Writes can be sent to the nearest cluster, reducing latency. Other options either use unsupported features or require manual replication, which is not ideal.

Answer

A single-cluster instance in us-central1 with a read replica in europe-west1.

Answer

Two separate Bigtable instances, one in us-central1 and one in europe-west1, with application-level replication.

Answer

A single-cluster instance in us-central1 with a Cloud Storage bucket in europe-west1 for backup.