Can you please provide advice on big data handling strategy?

Lucene is what Neo4j uses. (ref)


EDIT

Redis, which was a recent addition to SWI-Prolog, also does Full Text Search (ref)