Enabling an Enterprise Data Management Ecosystem using Change Data Capture with Amazon Neptune

Bradley R. Bebee, Rahul Chander, Ankit Gupta, Ankesh Khandelwal, Sainath Mallidi, Michael Schmidt, Ronak Sharda, Bryan B. Thompson, Prashant Upadhyay
2019 International Semantic Web Conference  
Given their flexibility in interlinking heterogenous data, graph databases are often used as a central hub within the enterprise data management ecosystem. While the data graph as such can be queried as an integrated data corpus using existing graph query languages (in Amazon Neptune, we support both SPARQL as a query language over RDF as well as Gremlin over the property graph data model), one key requirement of our customers is to integrate the data graph with external, purpose-built data
more » ... gement systems. In this demonstration, we will present Amazon Neptune's approach to synchronize graph data to external systems using Neptune's Change Data Capture (CDC) mechanism. We discuss the design and properties of the CDC feature and show how it can be used to synchronize graph data to external systems. Exemplified by a movie graph database which is periodically updated with new movies, we will showcase a faulttolerant cloud architecture leveraging CDC to periodically propagate updates made to the graph database into a backing Elasticsearch search index, in order to provide efficient full-text search over the graph data. On top of this stack, we demonstrate Neptune's approach to integrated querying across the data graph and the keyword search cluster.
dblp:conf/semweb/BebeeCGKM0STU19 fatcat:2ck5jtwzizbdxfo22ue54xo2xe