Hello, I hope I'm in the right category. My concern is about caching and requests.
I'm using neo4j Community 3.5.3 with a 4 million cells and 35 million relationships database (generated from CSV) to run shortest path algorithms (apoc astar and dijkstra).
I often resort to deleting the database (using rm -rf on the .db directory) and generating a new one using new CSVs.
How can I be sure the result to the request i'm running is not a cached one? (I often find almost identical results even if the graph weights are very different). Is it possible that a part of the nodes and relationship weights used in the computing is taken from cache, even after "rm -rf" deletion of the database? And eventually, is it possible that the database generates from a previous cached version of my input file? (I have read neo4j uses a file cache).
Thanks in advance !
