MaxCompute table data

The header of the table details page shows the table name, type tag, table description, and a group of buttons. Button visibility depends on your permissions, the current table type, and tenant activation:

• Request Permissions: Appears when your current account does not have access to this table. Click it to jump to Security Center and submit a permission request.

• Add to Album: Add the current table to a data album. If the table is already in an album, the button becomes an entry to view the albums it has joined.

• Favorite/Unfavorite: Add the current table to My Favorites, or remove it from favorites.

• API Generation: Publish the table's query capability as a DataService API. Visible only for supported data source types (such as MaxCompute, Hologres, AnalyticDB, PostgreSQL, MySQL, SQLServer, Oracle, OTS, SelectDB, MongoDB); not supported for tables in a lakehouse project.

• DataAnalysis: Jump to Data Analysis with a new query page automatically created for the current table. Supported on data source types such as MaxCompute (non-lakehouse) and public-cloud DLF.

• AI Enhancement: Use Copilot to generate or update the table's notes intelligently; while running, the button shows as Generating. Requires the AI enhancement permission and Copilot availability for your tenant.

• Refresh Metadata: Manually trigger metadata synchronization for the table; equivalent to the single-table action under My Data > My Tools > Refresh Table Metadata. Visible only for data sources such as MaxCompute, E-MapReduce, and public-cloud DLF in non-lakehouse projects.

The Table Basic Information section on the left side of the table details page displays information such as the number of Views, Number of Reads, and Number of Favorites.

• The

• Views: The number of times the table details page has been viewed in Data Map in the last 30 days. This data is collected offline and has a T+1 delay.

• Number of Reads: The number of tasks that read the MaxCompute table from the production environment in the last 30 days. Read tasks include SQL, Tunnel Download, Data Integration, and DataService Studio API calls. Currently, only reads from scheduled DataWorks tasks are counted. This data is collected offline and has a T+1 delay.

• Number of Favorites: The number of users who have added the table to their favorites. This data is collected in near real time.

• Output Nodes: The ID of the auto-triggered DataWorks node that writes to the current table. If the table is periodically updated but no node ID is displayed, a scheduled node outside DataWorks might be writing to it. Contact the table owner for details. This data is collected offline and has a T+1 delay.

  Note

  If you do not have permission to view the code of the output node, contact the administrator of the workspace where the node is located to grant the required permissions. For more information, see Enable code and log isolation for security.

• Storage Capacity: The logical storage size of the table. This data is collected offline and has a T+1 delay.

• Health Status: Displays the rating of the table's Governance Health Score, which is found in Data Health > Data Asset Governance. You can use this score to determine whether the table requires governance.

• Table Description: The description of the table, which Copilot can automatically generate.

If an administrator has configured custom attributes for table-type entities, a Custom Attributes card appears at the bottom of the left basic information panel, showing the current values of the configured attributes. Custom attributes can be inherited from the workspace; for more information, see Custom attributes.

Click Details to view the table's Field Information, Partition Key Column, and Change Records:

• Field Information

  You can view the field information of a table. If the table is partitioned, you can also view its Partition Fields.

  Actions	Description
  Edit	Click to edit the field's Description, Business Description, Security Level, and Primary Key. You can use Copilot to automatically generate field descriptions. Note Only a Workspace Administrator or the table owner can edit table fields. To grant this permission to other users, assign them the Workspace Administrator role. For more information, see Manage permissions for global services. The Security Level column is displayed only for tables where security levels are set for individual fields. You can set the security level for table fields only after you enable the field security level feature for the MaxCompute engine. For information about how to enable this feature, see Label-based access control.
  Recommended Field Desc	Generates descriptions for multiple fields that lack one.
  Batch Edit Security Level	Sets the security level for multiple table fields at once to enhance data security.
  Upload	Click this button and drag the local file that you want to upload into the Batch Upload Field Information dialog box. Note Only a Workspace Administrator or the table owner can upload data to the target table. To allow other users to upload data, grant them the Workspace Administrator role. For more information, see Manage permissions for global services. Only .xlsx files (Excel 2007 format) are supported. You can also Download Template File. This feature does not support model tables created in Data Modeling.
  Download	Click to download the field information for the current table.
  Generate SELECT	Click to view or Copy the select statement for the current table in the Generate SELECT Statement dialog box.
  Generate DDL	Click to view or Copy the table creation statement in the Generate DDL Statement dialog box.

  Note

  • Field access frequency: Shows the number of times the field was used in a JOIN clause in SQL on the previous day. The number is converted to a star rating, with a maximum of 5 stars and a minimum of 0 stars.

  • Associated Metric: Displays the model metrics associated with the field. To create or update the association, go to Dimensional Modeling, use field management on the editing page of the target table to maintain the field-to-metric association, and then publish the model table to apply the changes.

• Partition Information

  View the table's Partition Name, Number of Records, Storage Capacity, and other partition information.

  Note

  • The number of records and storage size of a partition are for reference only. Data updates may be delayed. The data in the compute engine is the source of truth.

  • For MaxCompute transactional tables, the Number of Records is not available and is always displayed as -1. The result of the SELECT COUNT(*) FROM <table_name> WHERE <partition>; command is the accurate value.

• Change records

  View the table's Description, Change Type, Granularity, and other change records.

  In the upper-left corner of the Change Records tab, you can select a change type from the drop-down list to view its specific change records.

Lineage information shows the relationships between tables and fields, derived from the actual data flow in operations such as job scheduling and data synchronization. On the lineage page, you can view the upstream and downstream nodes of tables or table fields, trace the original data sources, and see the final destinations of the data. You can also perform impact analysis across different lineage levels as needed.

• View Table Lineage

  

  On the Table Lineage tab, you can view the details of the current table's lineage. You can perform the following operations:

  • View the number of upstream and downstream nodes for each node in the lineage. Hover over a table or node type to see its basic information.

  • Click a node. In the pop-up panel under Lineage Association, enter a keyword to display all downstream nodes that contain that keyword. You can also enter @Username to display all downstream tables owned by the specified user.

  • Click the + or - icons in the lineage graph to expand or collapse the corresponding upstream and downstream nodes.

• View Field Lineage

  

  On the Field Lineage tab, you can view the details of a target field's lineage. You can perform the following operations:

  • Use the Change Field area to switch between fields of the current table and view the corresponding field lineage graph.

  • View the number of upstream and downstream nodes for each node in the field lineage. Hover over a field or node type to see its basic information.

  • Click the + or - icons in the lineage graph to expand or collapse the corresponding upstream and downstream nodes.

• Impact Analysis

  If a table's structure or data changes, its downstream nodes are affected. You can use Impact Analysis to identify which downstream tables a change will affect. On this page, you can filter by lineage level, node type, and table type to display relevant downstream tables and download the analysis results.

  Note

  Impact analysis supports up to 50 levels of table lineage.

You can view the data governance details and data quality status of a table.

• In the data governance details, you can view the table's governance health score, the trend of issues that require governance, and the details of governance issues. You can also quickly address the issues that need to be governed.

  • Health Score: A quantitative assessment of the table's health in five domains: storage, computing, development, quality, and security. The score reflects the effectiveness of governance on the table. A higher score indicates better governance.

  • Trend of to-Be-Governed Issues: Shows the trend of unresolved governance issues over time, helping you understand the table's governance history.

  • Governance Issue: Displays a list of current issues that require governance. You can click the issue name in the Governance Issue column or the action button in the Actions column to navigate to the Data Asset Governance > Overview > To-Do List page to handle governance issues.

• In the data quality details, you can view the details of the table's rules and any resulting data quality alerts.

This section displays the table's Usage Records, organized into Frequently Associated and Access Statistics.

• Frequently Associated: Shows how frequently the table is used in join operations.

  Note

  This statistic counts the number of times the table was used as a join condition in the last 30 days. This data is collected offline and has a T+1 delay.

• Access Statistics: Displays the table's usage records in chart format.

  • Trend for Reads: The date on the line chart corresponds to the number of reads on that day, distinguishing between reads from the development and production environments. The number of field associations is related to the number of node executions and the number of times the field appears in the code. This data is collected offline and has a T+1 delay.

    For example, if a field appears once in a node and the node is executed twice, the count is two. If a field appears twice in the code, a single node execution results in a count of two.

  • Field Popularity Details: Statistics on the number of times a field is used in SQL clauses (such as WHERE, SELECT, JOIN, and GROUP BY). This data is collected offline and has a T+1 delay.

  • Top 10 Readers: Statistics on users who have read the table in SQL within the last 30 days. This includes access from both production accounts for scheduling and personal accounts. The read operations include WHERE, SELECT, JOIN, and GROUP BY. This data is collected offline and has a T+1 delay.

Data Insight analyzes the structure and values of your data to display statistics and distribution information. You can create a new Data Insight report directly on the current page or go to the Data Analysis module to use Data Insight.

On the Output tab, view the list of DataWorks scheduled tasks that produce the current table. This helps you trace the data source and the most recent runs.

The list typically includes the task's Node ID, Node name, Owner, Schedule, and Latest status. Click the node ID to jump to the node details page in Operation Center, where you can further inspect task instances, schedule configuration, and run logs.

On the Usage Notes tab, view or maintain the usage notes of the table to help downstream users quickly understand its business meaning, column descriptions, and usage caveats.

• Read the notes: By default, the current notes are displayed in read mode.

• Edit the notes: Click Edit to enter edit mode. Depending on your tenant configuration, the editor can be the Yuque editor (rich text plus collaborative editing) or the DataWorks built-in editor.

• Generate notes with AI: Click the AI Enhancement button in the page header or use the entry inside the Usage Notes tab to let Copilot analyze the table metadata and produce a draft of the notes, which you can review and save.