What is the most efficient and fast way to load very large volumes of data into a Neo4j graph database?

What I'm trying to do right now is upload information for millions of people very quickly, efficiently and effectively in a graph database in Neo4j. Files containing data exceed 170 GB in size. Can you guide me by giving me an effective description of how I should go about achieving my goal? Obviously the person nodes needs to be connected with other nodes of different type. Thank you in advance!

Hi @kostas_vavouris !

Based on my experience, you can easily handle those as a CSV import.

The use of PERIODIC COMMIT is important in order to reduce the RAM load. Avoid trying to do everything in a single query. Try to guarantee the crreation of your base node (Person) first with a CREATE if possible. After it, go with the index creation and then the relationships.


Thanks! I will try your solution!