Principal mapping

Amazon Q Business uses principal mapping to map users and groups with permissions to access an Amazon Q Business application to their user ids and group membership information within the data sources that are connected to the application.

Although user and group mapping is a synchronous, simultaneous process, the following sections explain them separately for conceptual clarity.

User mapping

Each Amazon Q Business application can have multiple data sources connected to it. Each data source can have specific users and groups configured within it. Additionally, a user can be associated with multiple groups within a data source, or be attached to multiple groups across multiple data sources. A user attached to multiple data sources can also have different user IDs within these data sources.

A unique end user who signs in to an Amazon Q Business application must see only chat responses generated from documents that they have access to. To achieve that objective, Amazon Q Business maps their user IDs and group IDs within each data source to their identity provider (IdP) login credentials. Then, Amazon Q Business creates a universally unique identifier (UUID) to assign to each user. Using the UUID that it creates, Amazon Q Business stores a comprehensive mapping of the user’s group membership in an application. During chat, Amazon Q Business checks this UUID that's stored in its user store and retrieves user access information to generate chat responses.

The User Store feature also supports the following user management scenarios:

• An end user leaves your organization.

  When an end user leaves your organization, you can choose to delete the user from your user store.

• An end user leaves your organization, and their email gets recycled.

  Because User Store assigns each user a UUID for secure and accurate chat responses, email recycling doesn't impact the content that a user sees. Any new user within your application that's using a recycled email ID will be assigned a new UUID to be used for response generation.

• An end user with multiple login IDs needs chat content generated from documents they access using both these login IDs.

  With User Store, you can store user aliases attached to end user UUIDs. For example, a username Saanvi Sarkar uses two login IDs to sign in to Amazon Q Business—saanvi_sarkar and saanvi_s. You can store both IDs under the same UUID to ensure their chat responses are generated from content that they access using both login IDs.

Group mapping

Each Amazon Q Business application can have multiple data sources attached to it. Each data source in an Amazon Q Business application can have multiple groups attached to it. Multiple groups can repeat across multiple data sources. Additionally, each group across data sources can also contain multiple subgroups. Each Amazon Q Business application also has an associated identity provider (IdP) that can contain group information for the users accessing the application.

A unique end user signing in to an Amazon Q Business application must see only chat responses generated from documents within groups that they have access to. To achieve that objective, Amazon Q Business does the following:

• Automatically crawls local groups and their associated relationships from data sources during the connector configuration process.

• Provides you with API operations to map your end users group and subgroup membership details within each data source to their IdP group membership.

Then, Amazon Q Business creates a unique user identifier (UUID) to assign to each user. Under the UUID, Amazon Q Business stores a comprehensive mapping of the user’s group membership in an application. During chat, Amazon Q Business checks this UUID that's stored in its user store and quickly retrieves group access information to generate chat responses.

The User Store feature supports the following group management scenarios:

• Your users mapped to all groups that they have access to within an Amazon Q Business application.

  Amazon Q Business crawls all groups that a user has access to in a data source and stores this information under a user's UUID.

• Create a subgroup of users within your application.

  For example, for a group called company_employees, you might want to create a subgroup summer_interns and specify group level access for the subgroup. You might also want to group your interns into further subgroups like product_interns and engineering_interns.

• Map your data source groups to your IdP groups.

  A unique end user signing in to an Amazon Q Business application must see only chat responses generated from documents within groups they have access to. To support that objective, you can use Amazon Q to map your end users group membership details within each data source to their IdP group membership.

  Note

  Amazon Q Business doesn't interact or crawl this information from your IdP automatically.To ingest the relationship between data source groups and IdP groups, use the Amazon Q Business API.