All Products
Search
Document Center

DataWorks:Workspace management

Last Updated:Jul 19, 2025

A workspace is the basic unit for task development and member permission management in DataWorks. This topic describes workspace concepts, workspace modes, and the basic operations that you can perform in a workspace.

Workspace overview

  • All development operations must be performed in a custom DataWorks workspace. A workspace supports multiple types of data sources, such as MaxCompute, E-MapReduce, and real-time computing. After you create a data source and associate a compute engine with it in Data Development, you can develop tasks, schedule tasks, and manage data stored in the compute engine in the workspace.

  • The workspace administrator can add users to a workspace as members and assign different roles to the members to implement multi-role collaboration in different business scenarios. The roles include Workspace Administrator, Data Analyst, Deploy, Development, Model Designer, Visitor, O&M, and Security Manager.

  • Each workspace has completely independent member role settings and can be configured with different data sources. Therefore, you need to create a workspace before you start task development.

Workspace modes

DataWorks provides two types of workspace modes for users: basic mode and standard mode. The development behavior and data security risks of the two modes are different. We recommend that you select a workspace in standard mode for actual development. For more information, see Required reading: Differences between workspaces in basic mode and workspaces in standard mode and Scenario: Upgrade a workspace from the basic mode to the standard mode.

Basic operations

Operation

Description

References

Creates a workspace

Workspaces are basic units for managing tasks, members, roles, and permissions in DataWorks. All data development operations must be performed in a specific workspace.

Manage workspaces

Delete or disable a workspace

You can delete or disable a workspace.

  • If you delete a workspace, objects such as code in the workspace are also deleted.

  • If you disable a workspace, code that belongs to the workspace is retained but the workspace becomes unavailable.

View and modify the configurations of a workspace

After you create a workspace, you can view the configurations of the workspace and modify specific configurations. For example, you can upgrade the workspace mode.

Add data sources

You can add databases or compute engines to DataWorks by creating data sources. The data sources serve as the data sources for task execution in DataWorks. You need to create corresponding data sources for different environments based on the mode of your workspace.

Note

Before you create a data source, make sure that you are familiar with the differences between a workspace in basic mode and a workspace in standard mode. Subsequent data development and permission management operations vary based on the workspace mode.

Plan and assign roles for workspace members

  • After you create a workspace, you need to add RAM users to the workspace as members for collaborative development.

  • DataWorks lets you perform workspace-level permission management when you add RAM users. You can assign different roles to the RAM users to grant the RAM users the permissions on specific features.

Before you assign roles to the RAM users, make sure that you are familiar with the permission system for workspace-level service modules.

General guidelines for workspaces

When you create a workspace, we recommend that you follow these guidelines:

  • Uniform naming conventions:

    • Make sure that the workspace name contains information that helps identify your business.

    • Make sure that node names in a workspace contain the workspace name abbreviation for quick identification.

  • You can create a workspace using only an Alibaba Cloud account or a RAM user that is assigned the Workspace Administrator role. In principle, you should not assign both the Development and O&M roles to the same workspace member.

Appendix: Built-in logic of default workspaces

When you use DataWorks for the first time or activate DataWorks in a new region, DataWorks performs the following operations by default:

  1. Automatically generates a default workspace

    You can quickly start development operations without needing to make preparations.

  2. Automatically associates newly purchased resource groups with the default workspace

    After the association, you can directly use the exclusive resource group for scheduling and exclusive resource group for Data Integration in the default workspace and perform operations in the automatically built environment.

  3. Automatically initializes classic cases for the default workspace

    You can quickly get started with DataWorks by running these typical cases.

Important

We recommend that you do not develop tasks for the production environment in a default workspace.

Basic properties of default workspaces

The following table describes the default properties of DataWorks workspaces. In formal development scenarios, you can define the basic properties of workspaces. For more information about how to create a workspace, see Create a workspace.

Workspace property

Default workspace

Custom workspace

Workspace name

default_workspace_4-digit random code, which cannot be changed.

You can specify a name for a custom workspace. The name cannot be changed after the workspace is created.

Workspace display name

By default, the display name of a default workspace is Default Workspace. You can change the display name.

Custom. You can modify this after creation.

Workspace mode

By default, a default workspace is in basic mode. You can upgrade a default workspace from the basic mode to the standard mode. For more information, see Scenario: Upgrade a workspace from the basic mode to the standard mode.

You can create a workspace in basic mode or standard mode.

Workspace administrator

The workspace administrator varies based on the account that is used to activate DataWorks:

  • If you activate DataWorks using an Alibaba Cloud account, the workspace administrator of a default workspace is the Alibaba Cloud account.

  • If you activate DataWorks as a RAM user, the workspace administrator of a default workspace is the RAM user and the Alibaba Cloud account to which the RAM user belongs.

This is the same as the system default workspace.

Time zone for scheduling

By default, the time zone for scheduling is the local time zone. If you want to change the time zone, see Scenario: Switch the time zone for scheduling.

Same as the system default workspace.

Visibility of default workspaces

  • Tenant members can view and access only custom workspaces to which they belong. The roles of a tenant member in a custom workspace are assigned by the workspace administrator when adding the tenant member to the custom workspace.

  • All tenant members can view and access default workspaces. When a tenant member clicks a default workspace, they are automatically assigned the Visitor role. The workspace administrator can go to the Space Members Management page to reassign roles to workspace members.

Data sources and clusters associated with default workspaces

For a default workspace, the system automatically creates and associates a default MaxCompute compute resource. You can use this compute resource to run the cases provided by the platform and experience the features of DataWorks. For actual development, we recommend that you create your own workspace and associate the projects that are used for development operations, such as MaxCompute, E-MapReduce, and Hologres, with the workspace as compute resources. For more information, see Manage workspaces and Compute resource management.

Additional information: Start data development

After you create a workspace, add users to the workspace as members, and assign roles to the workspace members, the members can go to DataStudio to start data development based on the permissions of their assigned roles. For more information about how to perform data development operations in DataStudio, see Data development: Developers.

References