Amazon Neptune

Managed graph database — relationships as first-class data.

📖 Official AWS documentation ↗📰 Official AWS blog ↗

What is it?

A managed graph database supporting property graphs (Gremlin/openCypher) and RDF (SPARQL), with Neptune Analytics adding vector search over graph data for GraphRAG patterns.

💡 Why does it exist?

Some questions are ABOUT connections — fraud rings, knowledge graphs, recommendations by relationship — and joins in relational stores collapse under multi-hop traversals that graphs answer natively.

⏱️ When should you use it?

Use it for knowledge graphs grounding AI answers, fraud-ring detection, and identity graphs; consider GraphRAG when answers require traversing relationships, not just similar text.

🗺️ Where does it fit?

In the data layer beside your other stores: applications and ML pipelines traverse it via graph queries; Neptune ML integrates with SageMaker for graph neural network predictions.

🔌 How do you integrate it?

Create a cluster, load data in bulk from S3, query with Gremlin/openCypher/SPARQL; enable Neptune ML or Analytics when predictions or vector search over the graph are needed.

🧩 Commonly integrated with

Amazon S3Amazon SageMaker AIAmazon BedrockAWS Lambda

🎯 Exam angle (AIF-C01)

📚 Study it in a learning path

AIF-C01AWS AI PractitionerFlashcards, notes & quizzes covering this service →

More in Data & Analytics

Amazon S3AWS GlueAmazon AthenaAmazon OpenSearch Service