Graph DB

Why choose a graph db?

• You need to access data through multi-step traversals (find data in a network or hierarchy)
• You need flexibility on how to access the data in the future (a graph database is easier to change than other models)

concepts?

• Weighted relationships
• Can have bidirectional relationships
• Relationship pointing to itself (for modelling hierarchy?)

Graph Modeling Principles

1. Embrace the paradigm
   • Use the building blocks
     • Node, relationship, property
   • Anti-pattern
     • Rich properties -> instead normalize nodes
2. Nodes for Identity (37:00)
   • break out separate concepts
   • Property represents entity (I'm guessing it means turn the property into node(s))
     • (Reify | Nodify) connecting entities
     • simulating hyper-graph (relationship can have relationship) as Neo4j doesn't support it
3. Relationship for Access
   • Relationships for querying
     • Relationships should the primary means to access nodes in the database
     • Traversing relationships is cheap - that's the whole design goal of a graph database
     • Use indexes only to find starting node a query