Database Reference
In-Depth Information
In the different kinds of
properties
that can belong to the different elements of the
graph structure, the most basic ones, of course, are properties assigned to vertices
and edges.
An example of a simple property graph
Let's investigate this model in a bit more detail. Whenlookingcloseratthis,weind
the following interesting aspects of this model:
• Thereisnoixedschema.Thedatabase,inandofitself,doesnotimpose
that you have to have a schema, although most software professionals will
agree that having some kind of schema as you move closer to production is
probably not a bad idea.
• Partly because of the schema-less nature of the database, it seems to be a very
niceitfordealingwithsemi-structureddata.Ifonenodeorrelationshiphas
more or fewer properties, we do not have to alter the design for this; we can
just deal with that difference in structure automatically and work with it in
exactly the same way.
• Nodes and node properties seem to be quite easy to understand. In relational
terms, one can easily compare nodes with records in a table. It's as if the
property graph contains lots and lots of single-row tables, that is, the nodes
ofthegraph.Nodeswillhavepropertiesjustlikerecords/rowsinatable
willhaveields/columns.
• Relationships are a bit different. They always have a start- and an endpoint,
therefore have a direction. They cannot be dangling, but can be self-referencing
(same node as start- and endpoint). But the real power lies in the fact that:
