PCDE Manage a solution that can span multiple database technologies Practice Test 5 — 15 Questions

Question 1

A company is using Cloud SQL for MySQL and wants to perform disaster recovery testing by promoting a read replica to a standalone instance. Which TWO actions are required? (Choose 2)

Accepted Answer

Stop replication on the replica using the Cloud Console or gcloud.. Option A is correct because before a Cloud SQL read replica can be promoted, replication must be stopped so the replica is no longer applying changes from the primary; this is done via the Cloud Console or the gcloud sql instances promote workflow, which halts replication as part of promotion. Option E is correct because the actual promotion step is performed with the 'promote' action in the Cloud Console or the gcloud command (gcloud sql instances promote), which converts the read replica into a standalone primary instance that can accept writes. Option B is incorrect because you do not delete and recreate the replica; promotion is an in-place operation that preserves the instance and its data. Option C is incorrect because point-in-time recovery is a backup feature configured on the primary instance and is not a prerequisite for promoting a replica. Option D is incorrect because binary logging is used for replication and point-in-time recovery on the primary; disabling it on the replica is not required and would not enable promotion.

Answer

Delete the replica and recreate it as a primary instance.

Answer

Enable point-in-time recovery on the replica.

Answer

Disable binary logging on the replica.

Question 2

A company is migrating on-premises PostgreSQL databases to Cloud SQL. They need to minimize downtime and ensure data consistency. Which THREE steps should they follow? (Choose 3)

Accepted Answer

Use Database Migration Service to set up continuous replication from the on-premises database.. Option A is correct because Database Migration Service (DMS) supports PostgreSQL sources and performs continuous replication (change data capture) from the on-premises database to Cloud SQL, which is exactly what minimizes downtime during the migration. Option E is correct because the destination Cloud SQL for PostgreSQL instance must run a compatible/same major version as the source so that replication and schema/data compatibility are preserved. Option D is correct because once the DMS replication has caught up and the data is consistent, promoting the Cloud SQL replica to primary finalizes the cutover with minimal downtime. Option B is not appropriate because a dump/import is a one-time offline copy that causes downtime and does not provide continuous replication. Option C is not appropriate because a custom pg_dump/pg_restore script is manual, error-prone, and cannot maintain continuous consistency like DMS replication.

Answer

Export the on-premises database to a dump file and import it into Cloud SQL.

Answer

Manually sync data using a custom script with pg_dump and pg_restore.

Question 3

A company is using Cloud Bigtable and wants to set up monitoring and alerting for replication lag between clusters. Which TWO metrics should they use? (Choose 2)

Accepted Answer

cloudbigtable.googleapis.com/cluster/replication_lag. `replication_lag` directly measures the time difference between the primary cluster and a replica cluster in Cloud Bigtable, which is the key metric for monitoring replication delay. Option D is also correct because `replication_delay` is another metric that tracks the same concept, often reported in seconds, and is used to alert when replicas fall behind. Both metrics are essential for ensuring data consistency and timely failover in multi-cluster Bigtable deployments.

Answer

cloudbigtable.googleapis.com/cluster/disk_usage

Answer

cloudbigtable.googleapis.com/cluster/cpu_load

Answer

cloudbigtable.googleapis.com/cluster/operations_count

Question 4

A company runs an e-commerce platform on Cloud SQL for PostgreSQL. They need to perform point-in-time recovery (PITR) to recover from a user error that occurred 30 minutes ago. Which configuration is required to enable PITR?

Accepted Answer

Enable automated backups and configure a retention of 1-7 days; PITR uses WAL archiving automatically.

Answer

Enable binary logging and set a backup retention of 1-7 days.

Answer

Set up cross-region backup replicas and configure a retention of 1-7 days.

Answer

Enable PITR by setting the 'pitr_enabled' flag to true in the database flags.

Question 5

A Cloud Spanner database is experiencing increased read traffic. You need to add an index to improve query performance without downtime. What is the correct approach?

Accepted Answer

Use the CREATE INDEX statement. Spanner creates indexes online without blocking writes or reads.

Answer

Export the database, create the index in a new instance, then import.

Answer

Use the ALTER TABLE statement to add the index as a constraint.

Answer

Create a secondary index using the 'CREATE INDEX WITH OFFLINE' option to avoid blocking.

Question 6

You are monitoring Cloud Bigtable replication lag. Which metric should you use to determine if replicas are up to date, and what consistency level is typical for Bigtable replication?

Accepted Answer

Use 'replication_lag' metric; consistency is eventually consistent.. Cloud Bigtable exposes replication lag through the 'replication_lag' metric in Cloud Monitoring, which measures how far behind a replica cluster is relative to the primary. Bigtable replication is eventually consistent by default — writes propagate asynchronously across clusters, so reads from a replica may not immediately reflect the latest write. This combination of metric and consistency model is the standard Bigtable replication behavior.

Answer

Use 'cluster_lag' metric; consistency is read-your-writes consistent.

Answer

Use 'replica_lag' metric; consistency is strongly consistent.

Answer

Use 'replication_lag' metric; consistency is strongly consistent.

Question 7

A Memorystore for Redis instance is running out of memory. The application uses Redis as a cache with key expiration. You want to prevent data loss for keys that have not reached their TTL. Which eviction policy should you configure?

Accepted Answer

volatile-lru. The volatile-lru policy evicts the least recently used keys only from the pool of keys that have an expiration time set (TTL). This preserves keys without a TTL, which are typically persistent data that should not be lost. Since the application uses Redis as a cache with key expiration, volatile-lru ensures that only cache entries with TTLs are evicted, protecting any non-expiring keys from eviction.

Answer

noeviction

Answer

volatile-ttl

Answer

allkeys-lru

Question 8

A company has a Firestore database in Native mode. They need to run a query that filters on two fields (status and date) and orders by date. The query is slow and returns an error that a matching index is missing. What must the engineer do to resolve this?

Accepted Answer

Create a composite index on status and date in the Firebase Console or using gcloud.. Firestore requires a composite index on both the equality filter field (status) and the order field (date) when a query uses equality filters on one field and an order on another. Without this composite index, the query cannot be executed efficiently and returns an error. Creating the composite index via the Firebase Console or gcloud CLI resolves the issue.

Answer

Enable single-field indexes for both fields; Firestore will automatically use them.

Answer

Rewrite the query using 'IN' clauses to avoid the need for a composite index.

Answer

Change the database to Datastore mode, which does not require indexes.

Question 9

A team is designing a disaster recovery strategy for Cloud SQL. They need to be able to recover the database in a different region with a Recovery Point Objective (RPO) of less than 30 minutes. What should they configure?

Accepted Answer

Create a cross-region read replica and promote it to standalone during a disaster.

Answer

Use on-demand backups every 30 minutes and store them in a multi-regional bucket.

Answer

Enable binary logging and configure a cross-region replica for PITR.

Answer

Create automated backups with a retention of 7 days and restore to a new instance in the desired region.

Question 10

A Bigtable instance is experiencing high latency and uneven load distribution. The operations team suspects a hot spot. Which tool should they use to identify the hot spot, and what action can they take to mitigate it?

Accepted Answer

Use Key Visualiser to identify the hot row/key; then redesign the row key to distribute load.. Key Visualizer is the correct tool for identifying hot spots in Cloud Bigtable because it provides a heatmap of row key access patterns, allowing you to pinpoint specific rows or key ranges causing uneven load. Once identified, redesigning the row key (e.g., by salting or reversing keys) distributes writes and reads across nodes, mitigating the hot spot without requiring manual table splits or storage changes.

Answer

Use 'gcloud bigtable hot-spots list' command; then split the hot table into multiple tables.

Answer

Use Cloud Monitoring to check CPU utilisation; then add more nodes to the cluster.

Answer

Use the Bigtable admin console to view table statistics; then change the storage type from HDD to SSD.

Question 11

A Memorystore for Redis instance needs to be scaled to handle increased traffic. Which TWO methods can you use? (Choose 2)

Accepted Answer

Vertically scale by changing the tier to a larger machine type.. Option C is correct because Memorystore for Redis supports vertical scaling by changing the instance's tier or machine type (for example, from Basic to Standard or to a larger capacity tier), which increases memory and throughput for the instance. Option D is correct because Memorystore for Redis supports horizontal scaling via Redis Cluster mode, which shards data across multiple nodes to increase total capacity and throughput. Option A is incorrect because maxmemory is a Redis configuration parameter that limits usable memory and does not itself add capacity to a Memorystore instance. Option B is incorrect because changing the eviction policy (such as allkeys-lru) only affects how keys are evicted under memory pressure; it does not scale the instance. Option E is incorrect because Memorystore for Redis does not support adding user-managed read replicas in the same zone as a scaling method; read replicas are provided through the Standard tier's automatic replica, not by manually adding replicas.

Answer

Increase the maxmemory parameter.

Answer

Change the eviction policy to allkeys-lru.

Answer

Add read replicas in the same zone.

Question 12

A company is running a production Cloud SQL for PostgreSQL instance and wants to implement point-in-time recovery (PITR) with a 7-day retention window. They also need to ensure that automated backups are taken daily. Which two configurations must be enabled? (Choose TWO.)

Accepted Answer

Set the database flag 'archive_mode' to 'on' for WAL archiving. PITR requires WAL archiving, which is enabled by setting the 'archive_mode' to 'on' (or 'always') in the database flags. Automated backups are scheduled via the backup configuration in Cloud SQL. The retention period for PITR is set separately under backup settings.

Answer

Enable binary logging by setting 'log_bin' flag

Answer

Create a cross-region backup replica

Answer

Set the backup retention to 7 days using gcloud sql instances patch --backup-start-time

Question 13

An organization is using Memorystore for Redis and needs to ensure that when memory usage reaches the maximum, the cache evicts keys based on the least recently used (LRU) algorithm among keys with an expiry set. They also want to require password authentication for client connections. Which two configurations should be applied? (Choose TWO.)

Accepted Answer

Configure the AUTH password in the Memorystore instance. The eviction policy 'volatile-lru' evicts keys with an expiry set using LRU. AUTH is configured by setting a password via the Redis AUTH command or in the Memorystore instance settings.

Answer

Set maxmemory-policy to 'allkeys-lru'

Answer

Enable TLS for encryption

Answer

Enable persistence with AOF

Question 14

A developer is building an application that uses Firestore (in Datastore mode). The application needs to query data across two properties: 'status' and 'timestamp', with an equality filter on 'status' and a range filter on 'timestamp'. Which three steps are required to support this query efficiently? (Choose THREE.)

Accepted Answer

Ensure that single-field indexes exist for both 'status' and 'timestamp'. Firestore (in Datastore mode) requires single-field indexes to be defined for each property used in a query, even when a composite index is also present. Without a single-field index on 'status' and 'timestamp', the query engine cannot efficiently evaluate the equality and range filters, leading to full table scans or query failures.

Answer

Enable automatic composite index creation in Firestore settings

Answer

Create an index exemption for the 'timestamp' field

Question 15

A company is running a Cloud SQL for MySQL instance and needs to create a read replica in a different region for disaster recovery. They also want to be able to promote the replica to a standalone instance if needed. Which three steps should they take? (Choose THREE.)

Accepted Answer

Enable binary logging on the primary instance. Option A is correct because Cloud SQL for MySQL requires binary logging to be enabled on the primary instance so that the replica can receive and replay the primary's write operations via replication. Option C is correct because the gcloud sql instances create command with --master-instance-name and --region is the proper way to create a cross-region read replica of an existing Cloud SQL primary. Option D is correct because gcloud sql instances promote-replica is the documented command to promote a read replica to a standalone primary instance when a failover or DR event occurs. Option B is not required because Cloud SQL read replicas are read-only by design, so setting a database flag for read-only mode is unnecessary. Option E is incorrect because the scenario explicitly requires the replica in a different region for disaster recovery, not the same region as the primary.

Answer

Set the replica's database flags to enable read-only mode

Answer

Create the replica in the same region as the primary