Authorized datasets

This document describes how to use authorized datasets in BigQuery. An authorized dataset lets you authorize all of the views in a specified dataset to access the data in a second dataset. With an authorized dataset, you don't need to configure individual authorized views .

Overview

A view in BigQuery is a virtual table defined by a SQL query. For example, a view's query might return only a subset of the columns of a table, excluding columns that contain personal identifiable information (PII). To query a view, a user needs to have access to the resources that are accessed by the view's query.

Authorized views

If you want to let users query a view, without giving them direct access to the resources referenced by the view, you can use an authorized view . When you create an authorized view, you can share either a logical view or a materialized view. When you authorize a materialized view, it's referred to as an authorized materialized view .

An authorized view lets you, for example, share more limited data in a view with specified groups or users (principals), without giving the principals access to all of the underlying data. Principals can view the data you share and run queries on it, but they can't access the source dataset directly. Instead, the authorized view has access to the source data.

Authorized datasets

If you want to give a collection of views access to a dataset, without having to authorize each individual view, you can group the views together into a dataset, and then give the dataset that contains the views access to the dataset that contains the data. You can then give principals access to the dataset with the group of views, or to individual views in the dataset, as needed.

A dataset that has access to another dataset is called an authorized dataset . The dataset that authorizes another dataset to access its data is called the shared dataset.

Required permissions and roles

To authorize a dataset, or to revoke a dataset's authorization, you must have the following Identity and Access Management (IAM) permissions , which let you update the access control list of the dataset you are sharing.

After a dataset is authorized, you can create or update views in the authorized dataset. For more information and required permissions, see Create or update a view in an authorized dataset .

Permission	Resource
bigquery.datasets.get	The dataset you are sharing.
bigquery.datasets.update	The dataset you are sharing.

The following predefined IAM roles provide the required permissions.

Role	Description
bigquery.dataOwner	BigQuery Data Owner
bigquery.admin	BigQuery Admin

Quotas and limits

Authorized datasets are subject to dataset limits. For more information, see Dataset limits .

A dataset's access control list can have up to 2,500 total authorized resources, including authorized views , authorized datasets , and authorized functions . If you exceed this limit due to a large number of authorized views, consider grouping the views into authorized datasets. As a best practice, group related views into authorized datasets when you design new BigQuery architectures, especially multi-tenant architectures.

Authorize a dataset

You can authorize a dataset's current and future views to access another dataset by adding the dataset you want to authorize to the access list of the dataset you want to share, as follows:

Revoke a dataset's authorization

When you delete a dataset authorized to access another source dataset, it can take up to 24 hours for the change to fully reflect in the source dataset's access control lists (ACLs) . During this time:

• You won't be able to access the source data through the deleted dataset.
• The deleted dataset might still appear in the source dataset's ACL and count towards any authorized dataset limits. This could prevent you from creating new authorized datasets until the ACL is updated.

To revoke the access granted to the views in an authorized dataset, remove the authorized dataset from the shared dataset's access list, as follows:

Create or update a view in an authorized dataset

To create or update a view that is in an authorized dataset, you must have the permissions for the shared dataset that are listed in Required permissions and roles , in addition to the permissions that are required to create or update a view in a standard dataset.

The following table summarizes the necessary Identity and Access Management (IAM) permissions to create or update a view that is in an authorized dataset:

Permission	Resource
bigquery.datasets.get	The dataset you are sharing.
bigquery.tables.getData	Any tables or views from the shared dataset that are referenced in the new view you are creating or updating.
bigquery.tables.create	The authorized dataset in which you are creating a view.
bigquery.tables.update	The authorized dataset in which you are updating a view.

You don't need any additional permissions to delete a view from an authorized dataset.

Query a view in an authorized dataset

To query a view in an authorized dataset, a user needs to have access to the view, but access to the shared dataset is not required.

For more information, see Authorized views .

Authorized dataset example

The following example describes how to create and use an authorized dataset.

Assume you have two datasets, named private_dataset and public_dataset . The private_dataset dataset contains a table named private_table . The public_dataset dataset contains a view named private_table_filtered . The private_table_filtered view is based on a query that returns some, but not all, of the fields in the private_table table.

You can give a user access to the data returned by the private_table_filtered view, but not all of the data in the private_table table, as follows:

1. Grant the bigquery.dataViewer role to the user for the public_dataset dataset. This role includes the bigquery.tables.getData permission, which lets the user query the views in the public_dataset dataset. For information about how to grant a role to a user for a dataset, see Controlling access to datasets .

   The user now has permission to query views in the public_dataset , but they still cannot access the private_table table in private_dataset . If the user tries to query the private_table table directly, or if they try to access the private_table table indirectly by querying the private_table_filtered view, they get an error message similar to the following:

   Access Denied: Table PROJECT :private_dataset.private_table: User does not have permission to query table PROJECT :private_dataset.private_table.

2. In the BigQuerypage of the Google Cloud console, open the private_dataset dataset, click Sharing, and then select Authorize Datasets.

3. In the Authorized datasetpane that appears, enter PROJECT .public_dataset in the Dataset IDfield, and then click Add Authorization.

   The public_dataset dataset is added to the access control list of the private_dataset dataset, authorizing the views in the public_dataset dataset to query the data in the private_dataset dataset.

   The user can now query the private_table_filtered view in the public_dataset dataset, which indirectly accesses the private_dataset dataset, without having any permissions to directly access data in the private_dataset dataset.

Limitations

• You can create authorized datasets in different regions, but BigQuery doesn't support cross-region queries. Therefore, we recommend that you create datasets in the same region.

What's next

• For information about authorizing an individual view to access data in a dataset, see Authorized views .

• For information about authorizing a table function or a user-defined function to access data in a dataset, see Authorized functions .