Google Cloud

Google Cloud Certified – Generative AI Leader

PR000309
Trending

Become a certified Generative AI Leader with exam code PR000309 to enhance your AI skills in Google Cloud.

489 questions 0 views Free
Start Mock Test Timed · Full-length · Scored

Questions 311–320 of 489

Q311

A company needs to deploy a machine learning model in a cost-effective manner. What is the best approach?

  • A Use Google Cloud Run
  • B Use Google Compute Engine
  • C Use Google Kubernetes Engine
  • D Use Google App Engine
Explanation Google Cloud Run efficiently serves models in a serverless manner; others may incur higher operational costs.
Q312

What happens when a Google Cloud Storage bucket is deleted?

  • A Data is permanently deleted.
  • B Data is archived for recovery.
  • C Bucket is temporarily disabled.
  • D Data is replicated elsewhere.
Explanation Data in a deleted bucket is permanently gone; others suggest non-existent recovery options.
Q313

Which Google Cloud service is best for managing serverless APIs?

  • A Cloud Run
  • B Cloud Functions
  • C App Engine
  • D Compute Engine
Explanation Cloud Functions is specifically designed for serverless event-driven architecture, while Cloud Run and App Engine can also serve APIs but with more overhead.
Q314

You are configuring access for BigQuery datasets. What role provides complete access?

  • A BigQuery Data Editor
  • B BigQuery Job User
  • C BigQuery Admin
  • D BigQuery Data Viewer
Explanation The BigQuery Admin role grants all permissions, including full access to datasets, while other roles have limited permissions.
Q315

A machine learning model in Vertex AI has a decreased accuracy. What can you do first?

  • A Review deployment logs
  • B Increase the training dataset
  • C Re-train the model immediately
  • D Analyze feature importance
Explanation Analyzing feature importance can reveal if relevant features are missing or affecting accuracy, guiding informed adjustments, whereas the other options may not identify the root cause.
Q316

Which Google Cloud service is ideal for building machine learning models using big data?

  • A BigQuery
  • B Cloud Functions
  • C Cloud Pub/Sub
  • D Cloud Run
Explanation BigQuery is optimized for data analysis and ML model training, while others target different functions.
Q317

A company needs to manage user access to services based on their job role. Which feature should they implement?

  • A Cloud Identity
  • B Service Accounts
  • C IAM Roles
  • D VPC Peering
Explanation IAM Roles provide fine-grained access control, unlike the other options.
Q318

What happens when you enable AutoML on a dataset without enough labeled examples?

  • A Training will succeed with lower accuracy.
  • B Training will fail with a bad model.
  • C AutoML will estimate labels automatically.
  • D Training will succeed but be limited.
Explanation AutoML can train with limited data but results may be suboptimal.
Q319

Which Google Cloud service is designed specifically for building and managing machine learning models?

  • A Cloud ML Engine
  • B BigQuery
  • C Compute Engine
  • D Cloud Functions
Explanation Cloud ML Engine is tailored for machine learning tasks, while others focus on data processing or compute resources.
Q320

A company needs to ensure that only specific users can access certain datasets in BigQuery. What would be the most effective way to implement this?

  • A IAM roles at the project level
  • B Table-level access controls
  • C Query permissions only
  • D API key restrictions
Explanation Table-level access controls provide granular permissions specific to datasets, unlike project-level roles.