neo4j parallel traversal api -
the answer post (find distinct paths in neo4j database) recommends using traversal api identify distinct paths in graph (please see original question) instead of cypher query. graph containing 10,000 nodes api returned results immediately; faster cypher query.
i create same results graph containing more 100 mio. nodes. based on simple progress check added code, identifying distinct paths might take several hours. database files in total ca. 500 mb. memory not seem problem, cpu running @ 12% (on 8-core machine) => looks process cpu-constrained. there way make process run in parallel?
i using neo4j 2.1.3, jre 1.7.0_51-b13 (64-bit), runnning on windows 8.
many thanks!
Comments
Post a Comment