We are planning to migrate our inventory data from oracle to Neo4j . As per requirement we need to keep our both database in sync for some period in production to ensure about the correctness of application on Neo4j database.
We are using Apoc plugin to move one time data from oracle to Neo4j.
but our challenges are related with Real time data sync.
we are planning to use Neo4j data stream along with Oracle Golden Gate and Kafka . We are using Neo4j 3.5.5 version in our environment. Can you please confirm whether Neo4j stream available in 3.5.5 version and if not supported can you please guide us how to achieve this real time data sync approach?
Using a synchronization messaging approach like kafka is the right way to go.
If you need to purely write data to Neo4j from Kafka (sent by Oracle) I'd recommend that you look into the docs here, and use the Confluent worker plugin for neo4j-streams. This is a small bit of client code that lets you read messages off of kafka and send them to a remote Neo4j database, without using the Neo4j plugin. And yes it supports Neo4j 3.5 series
Exactly how to do it will depend on the nature of your data and how Oracle Golden Gate works, but in brief -- Oracle will be sending data on a number of different topics, and you'll write cypher queries to take the messages off of that topic and write them into graph patterns. Full details here: