Can I generate more than one distinct dump file? each one of them has diferent data filter from cluster
if not, I would use "neo4j-admin copy" with parameters
It is a good approximation to the problem?
Causal Cluster make de dump, and other standalone machines load the file, each machine has its peculiarities
Dump files are entire databases, and don't support projections or subsets. A graph is the graph. Post 3.5, the way we've handled these things is for people to partition into subgraphs with different databases, so you would keep different "profiles" if you will in different databases. You could then dump & load these individually as subsets/subgraphs.
But the only way to change what's in a dump file is to change what's in the database, because the dump is generally all or nothing.