All Products
Document Center

Configure Oracle full-database migration

Last Updated: Apr 04, 2018

Full-database migration is a convenient tool that can improve user efficiency and reduce cost by quickly uploading all the tables in an Oracle database to MaxCompute at a time. For more information about full-database migration, see Full-database migration overview.

This article demonstrates how to migrate a full Oracle database to MaxCompute by using the full-database migration feature.


  1. Log on to the Data Integration console, and select Offline Sync > Data Sources tab in the left-side navigation pane to enter the data source management page, as shown in the following figure.


  2. Click New Source in the upper-right corner to add an Oracle data source named clone_database for full-database migration, as shown in the following figure.


  3. Click Test Connectivity to verify whether the data source can be accessed correctly. Click Confirm to save the data source.

  4. After successful addition, the newly added Oracle data source clone_database is displayed in the data source list. Click Migration behind the Oracle data source to enter the full-database migration interface, as shown in the following figure.


    The full-database migration interface consists of three functional areas, as shown in the following figure.


    • Filter area of tables to be migrated: It lists all the database tables under the Oracle data source clone_database. You can select multiple database tables to be migrated.

    • Advanced settings: It provides the conversion rules of table names, column names, and column types between Oracle and MaxCompute data tables.

    • Control area of the migration mode and concurrency: You can select the full-database migration mode (full or incremental) and the concurrency (batch upload or full upload) and check the progress of submitting the migration task.

  5. Click Advanced Settings to select conversion rules as needed.

  6. In the control area of the migration mode and concurrency, select Daily Full as the synchronization mode.


    If the date field exists in your table, you can select Daily Incremental as the synchronization mode, and set the incremental field as the date field. Data Integration generates a where condition of incremental extraction for each task based on the selected incremental field by default and defines a daily data extraction condition by working with a DataWorks scheduling parameter such as ${bdp.system.bizdate}.

    To protect the Oracle data source from being overloaded by too many data synchronization jobs that start at the same point of time, Batch Upload can be selected. You can set to start synchronizing three database tables every one hour from 00:00 every day. At last, click the Submit Task button, and you can view the migration progress and the status of the migration tasks for each table.


  7. Click View Task corresponding to the table to jump to the task development page of Data Integration, and you can view the task running details.

    So far, you have completed migrating the full Oracle data source clone_database to MaxCompute. These tasks are scheduled to run according to the set scheduling cycle (daily scheduling by default). Also, you can transmit historical data by using the data completing feature of DataWorks. With the full-database migration of Data Integration, you can dramatically reduce the configuration for initial data migration to cloud and even the migration costs.