Moving to ingestion pipelines
In this chapter, we have created a static Neo4j graph, queried and analyzed it, and updated its features. This type of solution might be used in a production environment, where it might be used to serve up results in an application.
However, there are elements of this type of process that we are yet to cover. In a production system, nodes and edges may be read and written to a graph database regularly, often in small batches. Complicated processing might take place outside of Neo4j, before data is ingested, using Python or other languages. Think about how many options a real travel optimization or recommendation service actually offers – each of these has to be embedded into a graph database pipeline.
In Chapter 6, Pipeline Development, we will look at an example of a complex graph data pipeline and explore how to make it reliable and efficient.
