Amazon Neptune-to-OpenSearch replication

Amazon Neptune supports full-text search in Gremlin and SPARQL queries using Amazon OpenSearch Service (OpenSearch Service). You can use an Amazon CloudFormation stack to link an OpenSearch Service domain to Neptune. The Amazon CloudFormation template creates a streams-consumer application instance that provides Neptune-to-OpenSearch replication.

Before you begin, you need an existing Neptune DB cluster with streams enabled on it to serve as the source, and an OpenSearch Service domain to serve as the replication target.

If you already have an existing target OpenSearch Service domain that can be accessed by Lambda in the VPC where your Neptune DB cluster is located, the template can use that one. Otherwise, you need to create a new one.

We recommend that you use a newly created Neptune instance to use with OpenSearch Service. If you use an existing instance that already has data in it, you should perform an OpenSearch Service data sync before making queries or there may be data inconsistencies. This GitHub project provides an example of how to perform the synchronization: Export Neptune to OpenSearch (https://github.com/awslabs/amazon-neptune-tools/tree/master/export-neptune-to-elasticsearch).