Introduction to Amazon Neptune
Explore Amazon Neptune's purpose-built graph database capabilities, including property graph and RDF models. Understand how Neptune handles complex connected data queries efficiently with traversal languages like Gremlin and SPARQL. Learn when to choose Neptune over relational or document databases based on workload patterns such as multi-hop traversals, circular pattern detection, and connectivity scoring. This foundational lesson prepares you to design and work with graph databases on AWS.
When a query asks who knows someone who knows someone who bought this product, the database engine must traverse chains of relationships. In a relational system, each hop requires an additional JOIN operation, and the query planner must evaluate exponentially growing intermediate result sets. This is the exact problem space where graph databases deliver an architectural advantage. Rather than bolting relationships onto tables through foreign keys, a graph database treats relationships as foundational storage primitives, making multi-hop traversals a constant-time pointer chase instead of an expensive set intersection.
Relationship-centric workloads appear across many domains. Social networks query friendship chains. Fraud detection systems trace circular money flows through intermediary accounts. Recommendation engines walk purchase and preference paths. Identity resolution platforms link disparate records through shared attributes. Knowledge graphs classify entities within deep taxonomic hierarchies. In every case, the edges between entities carry as much analytical weight as the entities themselves.
Amazon Neptune is the AWS fully managed graph database service purpose-built for these workloads. It removes the operational burden of provisioning, patching, backups, and failover while supporting two distinct graph models and their associated query languages. This lesson establishes the foundational mental model you need before working with Neptune’s architecture. You will learn how property graphs and RDF differ, how core graph elements map to familiar relational concepts, and how to recognize scenario clues that point toward Neptune as the correct architectural choice.
The following diagram contrasts how a relational database and a graph database handle the same connected-data query.
With this visual contrast in mind, the next step is understanding the two graph models Neptune supports, starting with the property graph.
The property graph model
Property graphs are highly optimized for navigating complex relationships step by step.
Core elements of a property graph
A property graph stores data as a network of