Add a Replicat for Oracle AI Data Platform

Learn to add a Replicat for an Oracle AI Data Platform target in OCI GoldenGate.

Before you add a Replicat for Oracle AI Data Platform:

Ensure that the following privileges are added to the target:
- SELECT, CREATE, INSERT, UPDATE, and DELETE permissions on the target tables
- SELECT, CREATE, ALTER, and DROP permissions on the external tables
Create an OCI Object Storage connection and assign it to your Big Data deployment before configuring the Replicat.

To add a Replicat for Oracle AI Data Platform:

In the Add Replicat panel, on the Replicat Information page, complete the fields as needed, and then click Next:
- Select a Replicat Type:
  - Classic Replicat
  - Coordinated Replicat
- Enter a Process Name, no more than 5 characters long.
- Enter a Description, to help distinguish this process from others.
On the Replicat Options page, complete the fields as needed, and then click Next:
- For Replicat Trail:
  - Enter a Name.
  - Enter a Subdirectory, if you expect the Trail file to be in a location other than the default.
  - Select an Encryption profile. If an encryption profile wasn't created, the Local Wallet is selected by default.
  - For Begin, select the starting point for data processing:
    - Position in Log (default)
    - Now
    - Custom Time
- For Trail Position:
  - Enter the Sequence Number.
  - Enter the RBA Offset.
  - For Target, select Oracle AI Data Platform.
- Target Credentials:
  - Select a Domain.
  - Select an Alias within the specified Domain.
  - Select a Checkpoint Table. Leave the default selection or select the Checkpoint table you created for the target deployment.
  - Select the Max Threads Number (Coordinated Replicats only).
On the Managed Options page, complete the optional fields as needed, and then click Next:
- Select a Profile Name, if one is available.
- Enable Critical to deployment health.
- Enable and configure Auto Start settings.

On the Parameter Files page, you can add parameters to further configure your Replicat. For example:

If the MAP statement doesn't specify a target catalog, then the Oracle AI Data Platform workspace uses hive as the default catalog. For example:

MAP statement in Replicat parameter file	AI Data Platform Catalog	AI Data Platform Schema	AI Data Platform Table
`MAP SCHEMA_1.TABLE_1, TARGET "schema_1"."table_1";`	Default catalog, that is, `hive`	`schema_1`	`table_1`
`MAP DB_1.SCHEMA_1.TABLE_1, TARGET "db_1"."schema_1"."table_1"`	`db_1`	`schema_1`	`table_1`

If using Coordinated Replicat, add TARGETDB LIBFILE libggjava.so SET property=/u02/Deployment/etc/conf/ogg/your_replicat_name.properties to the parameter file.

On the Replicat Properties page, update the following properties. Ensure that you replace the placeholders with the compartment OCID and bucket name, respectively:
```
gg.eventhandler.oci.compartmentID=<oci-compartment-ocid>
gg.eventhandler.oci.bucketMappingTemplate=<oci-bucket-name>
```
Click Create and Run. If you click Create, then you can manually start the Replicat later from the Administration Service Home page or the Replicats page.

Oracle and/or its affiliates.