I have been working with Neo4j HA. It gives an average performance, which is also dependent on network and machines used. However, I've also noticed that its the write transactions, that take more time, thanks to the write transactions through the master. Now a part of my graph includes continuous reads and write operations.[Atleast 20 each, per second ]. I think this might be the reason for the average performance I'm getting. Is there any way, through which I can implement this sub-graph using in-memory DB of Neo4j?
Asked
Active
Viewed 182 times
3
-
Perhaps there is an answer for you, here? http://stackoverflow.com/questions/13325084/running-neo4j-purely-in-memory-without-any-persistence – Steven Graves Mar 08 '16 at 16:50