Differences between basic and standard mode workspaces-DataWorks(DataWorks)-阿里云帮助中心

Background

This topic covers the following aspects:

Section	Description
Introduction to basic and standard modes	Architecture of each workspace mode.
Impact on development and O&M	How each mode affects node development and O&M.
Pros and cons	Pros and cons of each mode.
Standard mode workflow	How standard mode enables role-based workflow governance.
Data source mapping for DataWorks modules in different workspace modes	Basic mode has only a production environment; standard mode has both development and production. This section maps DataWorks modules to their corresponding environments in each mode.
How to isolate development from production in basic mode	How to isolate development from production in basic mode.

Notes

In standard mode, you must create physically isolated data sources for the development and production environments. Create a data source.
Cross-project or cross-database access depends on data source characteristics. If the development and production environments use different data sources, whether the development environment can access production tables, resources, and functions depends on the data source type.
In standard mode, development environment tasks are not periodically scheduled by default. Tasks run on a schedule only after deployment to the production environment.

Key concepts

Basic and standard mode workspaces differ in the following ways:

Note

Either mode works for exploration, but standard mode is recommended for production use. It provides code isolation, compute resource isolation, permission isolation, and deployment workflow governance between environments.

To retain existing code while upgrading, see Upgrade a workspace from basic mode to standard mode.

The following table compares both modes in detail.

Perspective

Basic mode

Standard mode (recommended)

Number of data sources

A DataWorks workspace corresponds to one data source. Basic mode

A DataWorks workspace corresponds to two data sources, isolating the development and production environments.

Note

You must create physically isolated data sources for the development and production environments.

Standard mode workspace architecture

Corresponding DataWorks environments

The single data source serves as the production environment for DataWorks.

Of the two data sources, one serves as the development environment and the other serves as the production environment for DataWorks.

Note

You can create different data sources for the development environment and the production environment. For example:

Use different cloud instances for the development environment and the production environment.
Use different projects or databases within the same cloud instance for the development environment and the production environment.
In standard mode, if the data sources associated with the development environment and the production environment are different, running tasks in the development environment does not affect the production environment. To run tasks in the production environment, deploy the tasks to Operation Center and run them.

Impact on development and O&M

Comparison	Basic mode	Standard mode (recommended)
Development workflow governance	Submitted tasks are periodically scheduled in the scheduling system without deployment. (submit --> production)	Tasks must be submitted in the development environment and deployed to production before scheduling. (submit --> deploy --> production) Note In standard mode, only tasks in the production environment are automatically scheduled.
O&M permission governance	Developers can directly edit the code of production tasks.	Developers can only edit and submit code in Data Studio, not deploy directly to production. Deploying to production requires O&M permissions (project owner, workspace administrator, or operator role). Code can only be edited in the development environment, not in production. You can plan and govern the task development and O&M workflow in DataWorks based on the characteristics of a standard mode workspace and the DataWorks role-based permission system. For more information, see Standard mode workflow.
Data access governance	Developers can directly access production data for testing, posing security risks.	Developers use test data in the development environment, or request permissions to access production data for verification. Note Only MaxCompute supports requesting production data access through Security Center. MaxCompute data access control. Cross-project or cross-database access depends on data source characteristics. If the development and production environments use different data sources, whether the development environment can access production tables, resources, and functions depends on the data source type.
Data access identity	A unified identity is used to directly operate in the production environment. For MaxCompute, Hologres, EMR, and CDH, access identities include: Alibaba Cloud account, RAM user, RAM role (supported only by MaxCompute), and task owner. Note For other engines (AnalyticDB for MySQL, AnalyticDB for PostgreSQL, etc.), access depends on the account specified when the data source was created. Permissions match those of the database account.	Development environment: The default identity is Node Executor (the currently logged-in user) for task testing. Production environment: A specified identity is used to run scheduled tasks. You can go to Data Integration > Data Sources, find the target data source, and modify the access identity. Note MaxCompute, Hologres, EMR, and CDH Development environment: Task owner Production environment: Alibaba Cloud account, RAM user, or RAM role (supported only by MaxCompute) For other engines (AnalyticDB for MySQL, AnalyticDB for PostgreSQL, etc.), access depends on the account specified when the data source was created. Permissions match those of the database account.

Pros and cons

Comparison

Basic mode

Standard mode

Advantages

Simple and easy to use.

Assign the developer role and data developers can complete all data warehouse tasks.

Secure and standardized.

Provides a secure, standardized deployment workflow with code review and diff, ensuring production stability and preventing dirty data propagation or task errors from reaching production.
Data access is governed and security is enforced.

Disadvantages

Risks of instability and insecurity exist.

No environment isolation — development and production share the same data source, limiting you to simple data development.
You cannot control permissions on production tables.

Note
With MaxCompute, the developer role has read/write access to all tables in the project by default, posing data security risks.
You cannot govern the data development workflow.

Note
Developers can submit code to the scheduling system at any time without approval, introducing instability to production.

More complex workflow — typically requires multiple people to complete development and production tasks.

Standard mode workflow

The development-production isolation in standard mode affects data model design, processing logic, and code deployment workflows.

Appendix: Data source mapping by module

Go to DataStudio > Computing Resources to view compute resources associated with Data Studio. The following table shows which data source each module operates on:

DataWorks module	Standard mode	Basic mode
Data Studio	Operates on the development environment data source (instance, project, or database)	Operates on the production environment data source (instance, project, or database)
Operation Center	Development environment Operation Center: Operates on the development environment data source (instance, project, or database) Production environment Operation Center: Operates on the production environment data source (instance, project, or database)

Appendix: How to isolate development from production in basic mode

If you use a basic mode workspace and want environment isolation:

Prepare two basic mode workspaces — one for development, one for production. Use cross-workspace deployment to deploy tasks from the development workspace to the production workspace, achieving environment isolation. Cross-workspace deployment.

Drawback: The production workspace code can still be edited directly in Data Studio, so the production code entry point is not unique and may compromise workflow governance.

Recommendation: Upgrade to standard mode for better workflow governance. Upgrade a workspace from basic mode to standard mode.