Amazon Neptune
Amazon Neptune is a fast, reliable, fully managed graph database service that makes it easy to build and run applications that work with highly connected datasets. It supports property graph and RDF models, with multiple query languages including Gremlin, SPARQL, and openCypher.
APIs
Amazon Neptune Management API
Amazon Neptune Management API for creating, managing, and deleting Neptune DB clusters, instances, parameter groups, snapshots, and related infrastructure resources.
Amazon Neptune Data API
Amazon Neptune Data API provides SDK support for more than 40 data operations including data loading, query execution, data inquiry, and machine learning. It supports Gremlin an...
Neptune Gremlin API
Apache TinkerPop Gremlin graph traversal language API for querying property graphs in Neptune. It supports both WebSocket and HTTP REST endpoints for submitting Gremlin traversals.
Neptune SPARQL API
W3C SPARQL 1.1 query language API for querying RDF graphs in Neptune. It provides an HTTP REST endpoint compatible with the SPARQL 1.1 protocol specification.
Neptune openCypher API
openCypher graph query language API for querying property graphs with Cypher syntax in Neptune. It provides an HTTP endpoint for executing openCypher queries against property gr...
Neptune Streams API
Neptune Streams generates a complete sequence of change-log entries that record every change made to graph data as it happens, enabling real-time capture of graph mutations via ...
Neptune Loader API
Neptune bulk loader API for ingesting large volumes of data from Amazon S3 into a Neptune DB instance. It supports CSV formats for property graphs and multiple RDF serialization...
Neptune ML API
Neptune ML enables machine learning on graph data using graph neural networks. It provides APIs for data processing, model training, and inference endpoint management powered by...
Neptune Analytics API
Neptune Analytics is a memory-optimized graph database engine for analytics, providing optimized graph analytic algorithms, low-latency queries, and vector search capabilities w...
Capabilities
Amazon Neptune Analytics and Machine Learning
Workflow capability for Neptune Analytics graph analysis, vector search, and Neptune ML graph neural network model training and inference. Used by data scientists and ML engineers.
Run with NaftikoAmazon Neptune Graph Data Management
Workflow capability for managing Neptune graph databases, executing queries across Gremlin, SPARQL, and openCypher, and monitoring data streams. Used by graph database administr...
Run with NaftikoFeatures
Automatically scales compute and memory resources based on workload demands without requiring capacity planning.
Supports Apache TinkerPop Gremlin, openCypher, and SPARQL 1.1 query languages for property graph and RDF models.
Multi-AZ deployment with up to 15 read replicas, automated failover, and continuous backups with point-in-time recovery up to 35 days.
Multi-region replication with sub-second latency across up to five secondary clusters for global applications.
Memory-optimized graph analytics engine for analyzing tens of billions of relationships within seconds with vector search capabilities.
Fully managed GraphRAG with Amazon Bedrock Knowledge Bases for AI-enhanced graph retrieval augmented generation.
Native graph neural network support via Neptune ML powered by Amazon SageMaker for link prediction and node classification.
Full ACID transaction support ensuring data consistency and integrity across graph operations.
VPC network isolation, IAM resource permissions, AWS KMS encryption, TLS in-transit encryption, and CloudWatch audit logging.
Storage automatically grows up to 128 TiB with self-healing architecture spanning three availability zones.
Use Cases
Build knowledge graphs to enhance AI accuracy, comprehensiveness, and explainability using GraphRAG with Amazon Bedrock.
Model transaction and account relationship networks to detect fraudulent patterns in near real-time using graph traversals.
Build unified customer profile graphs linking purchases, preferences, and interactions for personalization and marketing.
Model IT infrastructure as a connected graph to detect attack paths, anomalies, and proactive threats.
Power product and content recommendation engines by traversing user-item relationship graphs.
Model and query highly connected social graph data for applications requiring relationship traversal at scale.
Map network topology, dependencies, and configuration relationships for operations and impact analysis.
Model complex supply chain relationships and dependencies for optimization and risk analysis.
Integrations
Integration with Bedrock Knowledge Bases for fully managed GraphRAG and AI-enhanced knowledge graph applications.
Neptune ML uses SageMaker for training graph neural network models on Neptune graph data.
Bulk data loading from S3 using the Neptune Loader API with CSV and RDF serialization format support.
Fine-grained resource-level access control and role-based permissions via AWS Identity and Access Management.
Metrics, logs, and audit logging for monitoring Neptune cluster performance and compliance.
Encryption at rest using AWS Key Management Service for customer-managed key support.
Network isolation using Amazon Virtual Private Cloud with security group and firewall controls.
Gremlin graph traversal language and TinkerPop ecosystem integration for property graph querying.
Integration with Strands AI Agents SDK and popular agentic memory tools for AI agent applications.