Table management

View table configurations

On the Table management page, find the table in the list and click View in the Actions column. The table name for index updates, the number of shards, and the change threshold for the index table are displayed on the right.

You can modify the index change rate as needed.

Click the drop-down arrow next to the table name. The detailed configuration of the index table appears below, including the configuration type and the version in use.

Edit a table

This section uses modifying the number of data shards as an example to demonstrate how to edit a table and apply the changes.

1. On the Table management page, find the table you want to edit and click Edit.

2. Edit the basic information for the table. Modify the number of data shards.

3. Data synchronization:

4. Change the index schema. You have two options:

   To keep the current schema: Select the no change option and click Next.To change the schema: Select any published version of the index schema to edit. You can edit the schema in form mode or developer mode.

5. Table configuration changes involve two scenarios: when changes are required and when they are not.

   To keep the current configuration: Select the no change option and click Next.

   To change the configuration, configure data processing and dictionaries:

   • Data processing configuration:

     By default, each data source includes two free resources. If you have a high volume of data updates and require high data timeliness, you can increase the value of `process_partition_count` in the data processing configuration to support larger updates.

     Modify the process_partition_count parameter:

     After you adjust this parameter, click Billing Document to view the billing details for the additional resources.

     After you finish editing and submit the changes, a new version is created.

   • Dictionary configuration:

     Dictionary configuration lets you customize tokenization. If the default tokenizer does not meet your business requirements, you can configure a custom dictionary for that tokenizer. This lets you influence the tokenization results.

4. Confirm your edits. Enter the required information based on the full data source, and then click Confirm.

   • MaxCompute data source: For the index rebuild method, you can choose Re-import full data or Rebuild current index data in-place.

   • Object Storage Service (OSS) data source: For the index rebuild method, you can choose Re-import full data or Rebuild current index data in-place.

   • Data Lake Formation (DLF) data source: For the index rebuild method, you can choose Re-import full data or Rebuild current index data in-place.

   • API data source: For the index rebuild method, you can choose Empty data, Recover data from index, or Rebuild current index data in-place.

     Important

     • Timestamp: Specifies the time range from which the new full version of the index retrieves incremental API data. The system can retrieve incremental API data from the last three days.

     • After you modify a table that uses a MaxCompute data source, the index rebuild process re-pulls the configured partition data. The process also pulls incremental API data based on the timestamp configuration.

     • The Empty data rebuild method deletes all previously pushed data and then starts to ingest real-time data from the specified timestamp. Use this option with caution.

     • The Recover data from index method requires you to select a data version to restore. For more information, see Recover data from index.

     • The Rebuild current index data in-place method builds the index using data from a selected version without re-importing the full data. For tables that use an API data source, this method prevents the data loss that can occur when you use the Empty data method. It also avoids the complex process of restoring data before you rebuild the index.

5. You can view the progress of the change in the change history.