starrocks data catalog - DataWorks - Alibaba Cloud Documentation Center

The data catalog provides a unified interface for managing StarRocks metadata. This topic describes how to create and manage data objects, such as tables and views, within the data catalog.

The StarRocks data catalog in DataWorks lets you:

Add existing StarRocks data sources to the catalog
Create and manage tables, views, and materialized views through a visual interface
Generate query SQL directly from the catalog tree
Remove data sources when no longer needed

Go to the StarRocks data catalog page

Go to the Workspaces page in the DataWorks console. In the top navigation bar, select a region. Find the workspace and choose Shortcuts > Data Studio in the Actions column.
In the navigation pane on the left, click the icon. In the Data Catalog tree, click StarRocks to open the StarRocks data catalog management page.

Create a StarRocks data catalog

On the StarRocks data catalog management page, add existing StarRocks data sources to the catalog list.

Prerequisites

Before you begin, make sure that:

A StarRocks data source has already been created in the current workspace. See Create a data source.
The StarRocks user has the SELECT privilege on the information_schema database. Without this privilege, metadata fails to load. See Manage users and data authorization.

Add data sources

Click the icon to the right of the StarRocks data catalog to open the add instance page.
On the DataWorks Data Source tab, add the StarRocks data source:
- To add a single data source, find it in the list and click Add in the Actions column.
- To add multiple data sources at once, select them and click Batch Add below the list.

Manage a StarRocks data catalog

In the StarRocks data catalog, manage data objects including tables, views, and materialized views.

Manage tables

Create a table

Click the icon next to the StarRocks project and find Table in the database under the catalog.
Click the icon to the right of Table to open the Create Table page.

Generate the basic table and field information using one of these methods:

Create a table with Copilot (recommended for new tables):
1. In the toolbar, click Create Table With Copilot to open the Copilot Chat interface.
2. Enter a natural language instruction, such as Create a daily user purchase details table.
3. Click Generate And Replace. Copilot generates a default table name and field information.
4. Click Accept to apply the result. Edit the generated information manually if needed.

Create a table manually:

You can create a table based on the parameter descriptions.

Parameter	Description
Basic information	Specify a custom Table Name, Table Description, and other details.
Field information	Add fields manually by clicking Insert and specifying the number of rows, field names, and types. Alternatively, click Generate Fields or Generate Field Descriptions to let Copilot generate them based on the table name and description.

Configure partition information. Choose a partition type based on your data volume and query patterns:

Partition type	When to use	Key parameters
No partition	Small data volume	None
Expression partition	Most scenarios — querying and managing data by continuous date ranges or enumerated values. Formerly known as automatic partition creation; more flexible and easier to use.	Expression Type, Time Function, Partition Field. See Expression Partitioning.
Range partition	Simple, continuous data such as time series data or continuous numerical data.	Range Type, Partition Field. See Data Distribution.
List partition	Data partitioned by explicitly defined enumerated values.	Partition Field, Partition Settings. See List Partitioning.

Configure advanced settings. Set the Sort Key and other parameters based on your table type. For more information, see the documentation on primary key tables, duplicate key tables, aggregate tables, and unique key tables.
Click Publish in the top toolbar to create the table.

Manage a table

View table details: Click the icon to the left of the StarRocks data catalog, then click Table to see basic information. Click a specific table name to view its Fields and DDL.
Generate query SQL: Right-click the table and select Generate Query SQL.
Delete a table: Right-click the table and select Delete.
Important
A deleted table cannot be recovered. Proceed with caution.

Manage views

Create a view

Click the icon next to the StarRocks instance and find View in the database under the catalog.
Click the icon to the right of View to open the New View page.
Follow the CREATE VIEW syntax. Replace view_name with your view name, and add the source query after the AS keyword.
Click Publish in the toolbar.

Manage a view

View details: Click the icon to the left of the StarRocks instance, then click View to see basic information. Click a specific view name to see its Fields and DDL.
Delete a view: Right-click the view and select Delete.

Manage materialized views

A materialized view pre-computes and stores query results, making it well-suited for complex queries over large tables — such as multi-table joins or aggregations — where recomputing results repeatedly would be expensive.

Create a materialized view

Click the icon next to the StarRocks instance and find Materialized View in the database under the catalog.
Click the icon to the right of Materialized View to open the Create Materialized View page.
Follow the CREATE MATERIALIZED VIEW syntax. Replace materialized_view_name with your view name, and add the source query after the AS keyword.
Click Publish in the toolbar.

Manage a materialized view

View details: Click the icon to the left of the StarRocks instance, then click Materialized View to see basic information. Click a specific materialized view name to see its Fields and DDL.
Delete a materialized view: Right-click the materialized view and select Delete.

View and remove a StarRocks data catalog

View the data catalog.
1. Click the icon to the left of the data catalog to see the added instances.
2. Hover over a StarRocks instance to see the name of the attached data source.
Remove a data catalog. Right-click the data catalog and select Remove.

FAQ

Why does metadata fail to load with the error "you need (at least one of) the SELECT privilege(s) on TABLE tables for this operation"?

Grant the SELECT privilege on the information_schema database to the StarRocks user. See Manage users and data authorization.