Oracle GoldenGate Integration

Oracle AI Data Platform integrates with Oracle GoldenGate to enable real-time replication of operational data into your AI Data Platform for analytics and AI-driven workloads.

With the integration of GoldenGate, you can continuously stream and merge transactional changes from source systems into AI Data Platform target tables.

For full details on configuration parameters, prerequisites, and troubleshooting, please refer to the official GoldenGate documentation.

Note:

Oracle GoldenGate documentation will be available with the next GoldenGate release in September.

How it Works

• GoldenGate captures change data from source databases and uses stage and merge data flow to enable high throughput, low impact data replication into AI Data Platform.
• GoldenGate can run initial instantiation and sync the initial load with cdc replication without any data loss. During initial instantiation or cdc replication, GoldenGate can automatically create target tables.

This allows AI Data Platform to always reflect the most up-to-date view of your operational data for downstream analytics, machine learning, and data engineering.

Requirements to Integrate with Oracle GoldenGate

• An OCI tenancy with an active AI Data Platform instance
• Database privileges on target AI Data Platform tables
• The SIMBA JDBC driver for Apache Spark, downloadable from the AI Data Platform console.

Configuration Summary

When you set up Oracle GoldenGate integration with Oracle AI Data Platform as part of the replicat process, you need to do the following:

• Select AIDP from the Target dropdown.
• Configure the replicat properties marked as TODO in the properties file.
• Set gg.target=aidp to enable AI Data Platform at the target.
• Specify a staging location. If no staging location is specified, OCI Object Storage is used.