Introduction to Neo4j

We could talk about the history and theory of graph databases, or we can dive in to the most popular graph database around. Let’s do the latter.

Getting Started

It’s easy to start. Just:

Download the latest version from neo4j.com. The Community Edition is freeeeeeeeeee.
Install it. On a Mac, open the .dmg file and do the drag-to-application-folder thing.
Run it. This starts a server at port 7474.
Go to http://localhost:7474 in a modern browser.
Login (and if it’s your first time, change your password).
Click on the Start Learning button and go through the mini-tutorial.

Review of the Fundamentals

Quick check. Did you remember these facts about graph databases in general?

A node represents an object.
A node has properties which are key-value pairs.
Property values can be numbers, strings, booleans, or lists of any of these three.
A node can have zero or more labels.
A relationship points from one node to another.
A relationship can have properties too.
You don’t use nulls! Just leave the property out!

The Neo4j browser allows you to type in queries and see the results as pictures or tables. The query language is called Cypher. Here are the first few Cypher queries from the tutorial. Remember what they did?

CREATE (ee:Person { name: "Emil", from: "Sweden", klout: 99 })

MATCH (ee:Person) WHERE ee.name = "Emil" RETURN ee;

CREATE (js:Person { name: "Johan", from: "Sweden", learn: "surfing" }),
    (ir:Person { name: "Ian", from: "England", title: "author" }),
    (rvb:Person { name: "Rik", from: "Belgium", pet: "Orval" }),
    (ally:Person { name: "Allison", from: "California", hobby: "surfing" }),
    (ee)-[:KNOWS {since: 2001}]->(js),
    (ee)-[:KNOWS {rating: 5}]->(ir),
    (js)-[:KNOWS]->(ir),
    (js)-[:KNOWS]->(rvb),
    (ir)-[:KNOWS]->(js),
    (ir)-[:KNOWS]->(ally),
    (rvb)-[:KNOWS]->(ally)

MATCH (ee:Person)-[:KNOWS]-(friends)
WHERE ee.name = "Emil" RETURN ee, friends

MATCH (js:Person)-[:KNOWS]-()-[:KNOWS]-(surfer)
WHERE js.name = "Johan" AND surfer.hobby = "surfing"
RETURN DISTINCT surfer

Cypher Basics

Cypher is a declarative language in which you write patterns. Variables in the patterns “retrieve” data when matched against your data graph.

. . . .

Why is Neo Cool?

Optimzied for fast traversal
Graphs can contain hundreds of trillions of entities
ACID
Schema optional, so no need for migrations
Conceptual model, design model, implementation model are all the same!
Query language is visual