What Is a Graph Database?

A graph database is one of the three major database types, along with relational databases and object databases. The main difference between a graph database and the other two is that graph databases function on relational graphing rather than using tables or mapping objects. There are three objects used in these databases: nodes, properties and edges. These databases are best for users who understand graphed information, and they move quicker at calculating associated data sets.

Graph databases are just that — databases that store information in a graphing format. Information is grouped based on whether it can be graphed together or not, and these databases are better for programmers looking to use graphing functions. When attached to websites, these databases are expensive, or use large amounts of code, to create the complicated database structure most websites need.

With a graph database, there are three parts, known as the nodes, properties and edges. The nodes represent entities. In a database that displays information about businesses and their relationship, the nodes would be the businesses. Each node is given a title, and the title is mostly so users and programmers can distinguish one node from the other nodes. The node’s title is rarely used in finding relations, but it can be used if the user needs.

The next piece of the graph database is the properties section. This goes along with the nodes and explains what the node is. Depending on what the database is graphing, the programmer will input relevant information. For the business database, if the programmer is looking to find similar businesses, the programmer will enter the business’s niche into the properties section; if the programmer is looking for similar sales figures, then the sales figures would be the properties.

Edges, the last section of the graph database, are lines that branch between the nodes and find relationships between the sections. For example, if the programmer wants to find similar businesses, the edges will graph between the nodes and find businesses that have matching properties. This gives users a visual graph on which he or she can see the relationship between the nodes.

The other two database types function in a similar way but have different ways of displaying or programming the databases. A relational database is nearly identical to a graph database, but it uses tables instead of graphs to show the similarities between database entries. Object databases use object-oriented programming (OOP) in which the objects function similarly to the nodes.