Practice Test 2 | Google Cloud Certified Professional Data Engineer | Dumps | Mock Test
Currently, anyone can access and modify the data sets, as well as creating & deleting data sets. The security team wants to restrict access of users on BigQuery and assign the minimum roles for each team based on their task requirements. You have the following teams:
-
-
- Data scientists: They should have read & write access to data sets. They may create & delete data sets.
- Data analysts: They have read access to data sets only.
- Development team: They need to create jobs to run queries for updating the website’s stats and product details.
-
What roles are recommended for each team?
A. Assign “roles/bigquery.dataOwner” role to data scientists. Assign “roles/bigquery.dataViewer” role to data analysts. Assign “roles/bigquery.user”role to the development team.
B. Assign “roles/bigquery.admin” role to data scientists. Assign “roles/bigquery.dataViewer” role to data analysts. Assign “roles/bigquery.dataViewer” role to development team.
C. Assign owner role to data scientists. Assign “roles/bigquery.dataViewer” role to data analysts. Assign “roles/bigquery.dataViewer” role to development team.
D. Assign admin role to data scientists. Assign “roles/bigquery.dataViewer” role to data analysts. Assign “roles/bigquery.user” role to development team.
Answer: A.
Here is the table of BigQuery roles and each role’s permissions:
From the table, we can assign each group the right role:
- Data scientists: They need to read & write to tables as well as creating and deleting datasets. So, owner & admin roles are the required roles. Since it’s best practice to grant the role with least permissions, we use owner role for this group.
- Data analysts: They need read access to data only. Viewer role is the permission allows them to read data from datasets.
- Development team: They need to create jobs on BigQuery. From the table, we see that user & jobUser roles may be suitable permissions. However, for developers, jobUser can be too tight for developers since they need to list tables and jobs. The answers list user role as a possible answer so we’ll stick to user role for this group.
Answer B is incorrect: admin role is too broad for data scientists. Viewer role doesn’t allow developers to create jobs.
Answer C is incorrect: Same as B, viewer role doesn’t allow developers to create jobs.
Answer D is incorrect: Same as B, admin role is too broad for data scientists.
Source(s):
BigQuery – Access Control: https://cloud.google.com/bigquery/docs/access-control
Comments are closed, but trackbacks and pingbacks are open.