Google Professional Data Engineer Question 62
Select 3Google Cloud PlatformYou are designing a data warehouse on BigQuery for your organization. The organization has strict data governance policies that require clear separation of data access between departments. Additionally, the organization requires monitoring of all data access activities and ensuring compliance with these policies. How should you design your project, dataset, and table architecture to meet these requirements?
- A
Create separate datasets for each department and assign IAM roles to control access at the dataset level.
- B
Use table-level access controls to restrict access to specific tables for each department.
- C
Enable BigQuery Audit Logs to monitor all data access activities.
- D
Create a single dataset for all departments and control access at the project level.
- E
Use Cloud Storage buckets instead of BigQuery to manage data and apply IAM roles at the bucket level.
Show answer and explanation
Correct answers: A, B, C
Explanation
To meet strict data governance requirements, it is important to design the data warehouse with clear separation of data (using separate datasets for each department) and granular access controls (using table-level permissions). Enabling BigQuery Audit Logs is necessary for monitoring and auditing data access activities to ensure compliance. A single dataset for all departments or using Cloud Storage instead of BigQuery would not meet the requirements for proper data governance in this scenario.
- A. Correct.
This is correct because creating separate datasets for each department is a best practice for ensuring clear separation of data and making it easier to manage access controls at the dataset level.
- B. Correct.
This is correct because using table-level access controls allows more granular control over who can access specific tables, which can enhance compliance with strict data governance policies.
- C. Correct.
This is correct because enabling BigQuery Audit Logs ensures that all data access activities are monitored, which is essential for compliance and governance requirements.
- D. Incorrect.
This is incorrect because creating a single dataset for all departments would make it harder to manage access controls and could lead to unintentional data exposure between departments.
- E. Incorrect.
This is incorrect because while Cloud Storage can be used for data management, it is not a replacement for BigQuery in a data warehouse design. BigQuery provides native features like datasets and table-level access controls that are more suitable for this use case.