A graph database is a database that prioritizes the relationships between records. Data is stored in a flexible schema made up of nodes, edges, and vertices. This model for storing information takes into account the native relationships between items as part of the schema, rather than the traditional method of establishing relationships through table joins and complex queries.
- Nodes are analogous to tables in a relational database. Each node represents a type of information stored in the database.
- Vertices are a unique entry within the node, the same as a relational database’s rows.
- Attributes describe each record within a vertex. For example, a vertex labeled “Customer” might have attributed like Customer Name, Phone Number, Credit Card Number, etc. In a relational database, attributes are the same as the columns in the table.
- Edges connect vertices by defining the relationship, similar to a table join in a relational database. For example, in the sentence “Customer resides at customer address,” the vertices “customer” and “customer address” are connected by the edge “resides at.”
Benefits of a Graph Database
- Graph databases make it possible to discover relationships and patterns in large data sets, opening the way for real-time analysis.
- Their lack of rigidity makes graph databases more flexible than the alternatives, allowing relationships to evolve as new data is introduced and making them more scalable and performant.
- Graph databases allow enterprises to build solutions for complex use cases, including machine learning and artificial intelligence solutions that weren’t possible using other database technology.
« Back to Glossary Index