All Products
Search

This Product

    Dataphin:Create a StarRocks compute source

    Document Center

    Dataphin:Create a StarRocks compute source

    Last Updated:Jun 23, 2026

    You can integrate StarRocks as an offline compute engine for Dataphin projects to process offline computing tasks.

    Background information

    StarRocks is a high-performance analytical database for real-time, multidimensional, and highly concurrent data analysis. It is highly extensible, available, and easy to maintain, and excels in OLAP scenarios such as real-time analysis, ad hoc queries, and data lake analytics. For more information, see the StarRocks official website.

    Permissions

    • All system roles except for tag business specialists and business members, and custom global roles with the Cluster-View permission, can view the details of each cluster.

    • Super administrators, system administrators, and custom global roles with the Cluster-Manage permission can create and manage StarRocks clusters. These users can also specify which users can reference the cluster when creating a StarRocks compute source and assign cluster administrators.

    • Cluster administrators can manage the clusters for which they are responsible.

    • Super administrators, system administrators, and users with the Compute Source Management-Create custom global role can create StarRocks compute sources. They can also select and reference StarRocks clusters that they have permission to use.

    Procedure

    1. On the Dataphin home page, choose Planning > Compute Source from the top menu bar.

    2. On the Compute Source page, click Add Compute Source and select StarRocks Compute Source.

    3. On the Create StarRocks Compute Source page, configure the following parameters.

      Reference a specified cluster

      Parameter

      Description

      Basic Information

      Compute Type

      Set to StarRocks.

      Compute Source Name

      The name can contain Chinese characters, letters, digits, underscores (_), and hyphens (-).

      Configuration Method

      Select Reference A Specified Cluster. From the drop-down list, select a cluster that the current user has permission to reference. Click View to go to the View StarRocks page for cluster details. If the required cluster is not available, click Configure Cluster to go to the Create StarRocks Cluster page and create a new one.

      Note

      Changes to the cluster information are automatically synced to the current compute source.

      Compute Source Description

      Enter a brief description for the compute source. The description can be up to 128 characters long.

      Configuration

      JDBC URL

      The JDBC URL from the selected StarRocks cluster is used by default and cannot be modified.

      Catalog

      Select Default Catalog or External Catalog.

      • Default Catalog: Manages internal data in StarRocks.

      • External Catalog: Lets you select from all external catalogs under the cluster or manually enter a catalog name.

      Database

      Select a database from the selected catalog, or manually enter a database name.

      Authenticated User

      You can select Same As Cluster or Custom. The default is Same as cluster. If you select Custom, enter the Username and Password. Make sure the user has the required database permissions.

      Task Resource Group

      StarRocks uses resource groups to isolate resources and classifiers to match tasks to resource groups. You can specify resource group names for tasks with different priorities to ensure automatic assignment at runtime.

      • Use Default Execution User: Uses the default execution user configured in the selected cluster.

      • Custom: Includes five priority levels: Highest Priority, High Priority, Medium Priority, Low Priority, and Lowest Priority. You must enter a resource group name for each priority level.

      Manual configuration

      Parameter

      Description

      Basic Information

      Compute Type

      Set to StarRocks.

      Compute Source Name

      The name can contain Chinese characters, letters, digits, underscores (_), and hyphens (-).

      Configuration Method

      Select Manual Configuration.

      Compute Source Description

      Enter a brief description for the compute source. The description can be up to 128 characters long.

      Configuration

      JDBC URL

      Enter the JDBC URL. The following formats are supported:

      • jdbc:mysql:loadbalance://{fe1-host}:{port},{fe2-host}:{port},{fe3-host}:{port}/{database}

      • jdbc:mysql://{host}:{port}/database?key1=value1&key2=value2

      Catalog

      Only Default Catalog is available. To use an external catalog, select Reference A Specified Cluster as the configuration method instead.

      FE Node URL

      The connection addresses of Front End (FE) nodes. Separate multiple addresses with commas (,). Example: fe_host1:http_port01,fe_host02:http_port02.

      Username

      Enter the username and password for the StarRocks database. Make sure the user has the required database permissions.

      Password

      Task Resource Group

      StarRocks uses resource groups to isolate resources and classifiers to match tasks to resource groups. You can specify resource group names for tasks with different priorities to ensure automatic assignment at runtime.

      • Use Default Execution User: Uses the default execution user from the preceding configuration.

      • Custom: Includes five priority levels: Highest Priority, High Priority, Medium Priority, Low Priority, and Lowest Priority. You must enter a resource group for each priority level.

    4. Click Test Connection to test the connection to the compute source.

    5. After the connection test is successful, click Submit.

      After you create the StarRocks compute source, you can attach it to a project. For more information, see Manage permissions and compute sources for a project space.

    What to do next

    After you create the StarRocks compute source, bind it to a project. For more information, see Create a general project.