Our open source initiative is populating a Google Firestore from an XML data source. Afterwards we'd like to program a way to setup a Neo4j database and synchronize the data across the two databases. Mainly we just want to see how the data transfer and synchronization could work and see what sorts of challenges arise. Any thoughts on the setup?
I'm just getting into neo4j.
We sync our firestore dbs to MongoDb and others, but it seems that neo4j lacks some features like handling maps, so there's no way to look for an document.mymap.mykey === myvalue , It could be stored as a serialized object so it will not be available on searches. If you update arrays on real time, some mechanism to avoid race conditions should be set.
I have some other doubts like how far can the system go on storage holding billion of nodes and relationships. Maybe the system relies all in memory. Oh! Last thought is that all denormalization should be left off the nodes.