Database Reference
In-Depth Information
These types of index structures are very often used to deal with time data in a graph
structure. In this case, we create more of a time tree instead of a time line and connect
our graph data to this tree instead of putting timestamps as properties on every node
or relationship. The following diagram contains an example of such a tree structure:
An in-graph time-tree index
All of the patterns in the preceding diagram are common modeling patterns that have
been used successfully in projects. Use them wisely, and always remember that it's all
about the query patterns. Knowing what the questions you want to ask of the data are
will massively impact its design model, and chances are that you will need and want
to iterate over that model multiple times to get it to a stable state. Fortunately, graph
database management systems such as Neo4j deal with this kind of variation quite
elegantly and allow you to do exactly this when appropriate.
