Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Professional Cloud Architect practice bank has 2,335 more like it. Try 10 free: instant score, no account.
An enterprise maintains a multi-project architecture on Google Cloud. Raw source data is stored in BigQuery tables within a dedicated data warehouse project (project-data), while the machine learning platform team manages a Vertex AI Feature Store instance in a separate project (project-ml).
You need to configure access control so that the Vertex AI Feature Store in project-ml can ingest batch feature data from BigQuery in project-data. In addition, your organization's security policy requires that IT/DevOps administrators who configure and scale the Feature Store infrastructure cannot read or write actual feature values.
Which identity and access management (IAM) configuration should you implement?
project-ml the roles/bigquery.admin role in project-data, and assign the roles/aiplatform.featurestoreAdmin role to the IT/DevOps administrators in project-ml.This approach grants the Compute Engine default service account broad administrative access to BigQuery in the source data project and assigns the full roles/aiplatform.featurestoreAdmin role to the DevOps team.
@gcp-sa-aiplatform.iam.gserviceaccount.com)—not the Compute Engine default service account—to perform backend ingestion tasks from BigQuery.roles/aiplatform.featurestoreAdmin role includes full read and write permissions over featurestore contents (aiplatform.features.* and data operations), directly violating the policy that prevents IT/DevOps from reading or writing feature values.DevOps engineers were required to manage the entire data lifecycle including schema updates, feature value modification, and ad-hoc data deletion, and if workloads ran inside self-managed Compute Engine virtual machines.
roles/aiplatform.featurestoreDataWriter role to the IT/DevOps administrators in project-ml, and grant the Vertex AI Service Agent from project-data read permissions in project-ml.This configuration assigns data-writing privileges to IT/DevOps administrators and attempts to authorize the service agent belonging to the data project (project-data) instead of the consuming project.
roles/aiplatform.featurestoreDataWriter role is meant for data engineers and automated pipelines that write and ingest feature values into the store. It violates the constraint that DevOps administrators must not have data write or modification access.project-ml, which means the service agent residing in project-ml needs permissions on project-data, not vice-versa.roles/aiplatform.featurestoreDataWriter grants write access to entity data and features rather than infrastructure instance management.The DevOps team were responsible for creating and writing feature values directly, and the featurestore resided in project-data pulling data from project-ml.
roles/aiplatform.user role to IT/DevOps administrators in project-ml, and configure a VPC Network Peering connection between project-data and project-ml to bypass IAM authorization.This option suggests using VPC Network Peering to bypass IAM checks for BigQuery data ingestion while granting the broad roles/aiplatform.user role to the infrastructure team.
roles/aiplatform.user role allows users to create and manage various AI workloads and resources without restricting data-level access to feature values in the manner required by the security policy.The workload involved private Compute Engine VM-to-VM communication over internal IP addresses rather than Google Cloud managed API data integration.
service-PROJECT_NUMBER@gcp-sa-aiplatform.iam.gserviceaccount.com) from project-ml read access to the BigQuery datasets in project-data, and assign the roles/aiplatform.featurestoreInstanceCreator role to the IT/DevOps administrators in project-ml.Vertex AI Feature Store acts on behalf of users to execute background tasks—such as batch ingestion and feature serving—using a Google-managed service agent with the format service-PROJECT_NUMBER@gcp-sa-aiplatform.iam.gserviceaccount.com. In addition, Vertex AI provides fine-grained predefined IAM roles designed for distinct organizational personas, including administrators, data scientists, and DevOps personnel.
project-ml). To import batch data across project boundaries from project-data, the service agent of project-ml must be explicitly granted read permissions (such as roles/bigquery.dataViewer and roles/bigquery.jobUser) on the source BigQuery datasets.roles/aiplatform.featurestoreInstanceCreator role allows IT/DevOps engineers to provision, tune, and maintain featurestores without granting them read or write permissions to the sensitive feature data stored inside.
This pattern establishes standard least-privilege principles by separating infrastructure operations from data payload access and delegating cross-project ingestion authority strictly to the Google-managed Vertex AI service agent.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated