PCD Practice Test 9 — 10 Questions

Question 1

A manufacturing company collects sensor data from thousands of devices. They need a database that can handle a write throughput of 100,000 rows per second and read latency under 10ms. Data is keyed by device ID and timestamp. Which Google Cloud database should they choose?

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

Accepted Answer

Cloud Bigtable. Cloud Bigtable is a wide-column NoSQL database designed for high-throughput, low-latency workloads with time-series data keyed by device ID and timestamp. It can handle 100,000 writes per second and sub-10ms read latency, making it the correct choice for this IoT sensor data scenario.

Answer

Cloud Spanner

Answer

Cloud SQL

Answer

Firestore

Question 2

A company needs to back up their Cloud Bigtable instance for disaster recovery across regions. They want to restore to a different cluster in a different region. What should they use?

Accepted Answer

Use Bigtable managed backups and restore to a different cluster. Cloud Bigtable managed backups are the native, supported mechanism for disaster recovery and can be restored to a different cluster, including one in a different region, making them the correct choice for cross-region restore. Managed backups capture table data and schema and are designed exactly for this DR scenario. This avoids the complexity and consistency risks of manual export/import workflows.

Answer

Export tables to Avro in GCS and import in another region

Answer

Use gcloud bigtable instances create with replication

Answer

Use Cloud Scheduler to copy tables via Dataflow

Question 3

A company uses Cloud SQL for MySQL. They need to export data to Cloud Storage regularly. What is the recommended method?

Accepted Answer

Use Cloud SQL export feature to export to Cloud Storage.. Using the Cloud SQL export feature to export to Cloud Storage is correct because it is the native, recommended method for exporting data from Cloud SQL to Cloud Storage. It handles the export in a managed way, supports CSV and SQL formats, and integrates with IAM and Cloud Storage permissions. This method is efficient and requires minimal setup.

Answer

Use Dataflow to read from Cloud SQL and write to Cloud Storage.

Answer

Use a cron job on Cloud SQL instance to write to Cloud Storage.

Answer

Use mysqldump command from a Compute Engine instance.

Question 4

An application running on Compute Engine uses Cloud Storage for storing user-uploaded images. During load testing, the application experiences high latency when reading images. The developer suspects that the application is making too many small read requests. Which approach should the developer take to optimize performance?

Accepted Answer

Increase the read size to reduce the number of API requests.. The high latency is caused by many small read requests, each incurring API overhead. By increasing the read size (e.g., reading larger chunks or using range requests), the application reduces the number of API calls, which lowers cumulative latency and improves throughput. This directly addresses the root cause of excessive small reads.

Answer

Enable Cloud CDN to cache the images at edge locations.

Answer

Rewrite the objects to use a different storage class.

Answer

Mount the Cloud Storage bucket using Cloud Storage FUSE and read files from the local filesystem.

Question 5

You are designing a Cloud Spanner schema for a global user database. Users are identified by a UUID. You notice hotspotting on a specific node during writes. What is the most effective way to resolve this?

Accepted Answer

Use a composite primary key with a hash prefix of the UUID. Hotspotting in Cloud Spanner occurs when writes are concentrated on a single node due to a monotonically increasing primary key (like a UUID without a hash prefix). By using a composite primary key with a hash prefix of the UUID, you distribute writes evenly across all nodes because the hash prefix ensures that successive writes land on different splits. This prevents a single node from becoming a bottleneck, thereby resolving the hotspotting issue.

Answer

Enable interleaved tables to store child rows with parent

Answer

Change the primary key to a monotonically increasing timestamp

Answer

Add a secondary index on the UUID column

Question 6

Which TWO statements about Cloud Trace are correct?

Accepted Answer

Trace can be integrated with Cloud Monitoring for alerting. Option A is correct because Cloud Trace integrates with Cloud Monitoring, allowing you to create alerting policies based on trace-derived metrics such as latency distributions and error rates, so you can be notified when performance degrades. Option D is correct because Cloud Trace is specifically designed to collect and correlate spans across distributed services, enabling end-to-end latency analysis of requests as they flow through microservices and other components. Option B is not correct because Trace does not collect latency data from all requests by default; sampling is applied (often 0.1 requests per second per instance plus a percentage of additional requests) unless you configure custom sampling or use the Trace API/SDK to instrument and control what is captured. Option C is not correct because Trace does not automatically create dashboards; visualization of trace data is done in the Trace console, and dashboards are created in Cloud Monitoring, not automatically by Trace. Option E is not correct because auto-scaling based on latency is a function of services like Managed Instance Groups or GKE autoscaling (often using Monitoring metrics), not a capability of Cloud Trace itself.

Answer

Trace collects latency data from all requests by default

Answer

Trace automatically creates dashboards for visualization

Answer

Trace supports auto-scaling based on latency

Question 7

A company deploys a containerized application to Cloud Run using Cloud Build. They want to implement a rolling update strategy with zero downtime. Which two actions should they take? (Choose two.)

Accepted Answer

Gradually shift traffic to the new revision using the gcloud run services update-traffic command.. The `gcloud run services update-traffic` command allows you to gradually shift traffic from the current revision to a new revision, enabling a rolling update with zero downtime. This command supports percentage-based traffic splitting, which ensures that the new revision is incrementally exposed to users while the old revision remains active, thus maintaining service availability throughout the deployment.

Answer

Create a new Cloud Run service for the new revision.

Answer

Set the min-instances attribute to 1 to keep at least one instance running.

Answer

Use the gcloud run deploy command with the --concurrency flag.

Question 8

A developer sets environment variables for a Cloud Function as shown. What is the security concern?

Accepted Answer

The password is exposed in plain text and should be stored in Secret Manager.. Storing a database password directly in a Cloud Function's environment variables (e.g., DB_PASS=mysecretpassword) exposes it in plain text. Environment variables are visible in the function's configuration, logs, and deployment metadata, and anyone with read access to the function can see the secret. Google Cloud recommends using Secret Manager to store sensitive data and referencing it at runtime, which provides encryption, access control, and audit logging.

Answer

The variable name FOO is too short and does not follow naming conventions.

Answer

The variable DB_PASS should be set as a build variable instead.

Answer

There is no encryption applied to environment variables.

Question 9

A team is migrating a 500 GB Snowflake data warehouse to BigQuery. They have a tight deadline and need to minimize manual SQL rewriting. Which approach is most efficient?

Accepted Answer

Export Snowflake data to Cloud Storage as Parquet, create external tables, and use BigQuery's SQL translation for queries.. The most efficient approach to minimize manual SQL rewriting is to export Snowflake data to Cloud Storage as Parquet, create external tables, and use BigQuery's SQL translation tool. This tool automatically converts Snowflake-specific SQL syntax to BigQuery-compatible syntax, significantly reducing manual effort. Option C (Data Transfer Service) does not perform SQL translation; it only handles data movement. Option A (Cloud Data Fusion) requires custom transformations and does not address SQL rewriting. Option B is manual and time-consuming.

Answer

Use Cloud Data Fusion to transform Snowflake data into BigQuery-compatible format.

Answer

Manually rewrite all SQL queries to BigQuery syntax using a text editor.

Answer

Use the BigQuery Data Transfer Service with Snowflake connector.

Question 10

A mobile app needs to store user preferences as a simple key-value store. The data is infrequently accessed and should be highly available with no operational overhead. The app is already using Firebase. Which database is the best fit?

Accepted Answer

Firestore. Firestore is serverless, integrates with Firebase, and is ideal for storing user preferences as documents.

Answer

Bigtable

Answer

Memorystore

Answer

Cloud SQL