Database Reference
In-Depth Information
Figure 5-2.
Graph of all users following all other users
Adding the Content Using a Linked List
The next step in the process is loading content and relating them to Users. For the purposes of this example, I generated
three separate CSV files that we are using to create a linked list of status updates, rather than relating each specific status
update directly to a User. Although the number of status updates is limited to three in this example, in a real-world
application the expectation is that the number of status updates will grow.
Indeed, some users may add thousands of status updates, so the number of direct relationships could grow into a
densely connected Node. The latest release of Neo4j helps with the dense node problem by splitting relationships by
type and direction, which will help as Users take on more followers and follow other Users.
However, it also makes sense to address this from a graph perspective to improve performance and design. In
most cases, the status updates will be returned in pages of a specific value—say, 15 per page—and the application will
not require immediate access to each connected status update other than through an identifier to show a single status.
In addition, the retrieval and filter of nodes can happen with the sequence of nodes already in their desired order.
