I think Michael's question is still valid here: What do you intend to 
achieve by using an in-memory graph? Of course if the available memory is 
limited as compared to the volume of graph data, any in-memory solution 
will have to resort to some kind of disk-based paging once memory is full. 
If what you are trying to solve is performance issues, there are ways to 
make Neo4j keep everything in memory, but that needs ample memory and some 
customized handling of caching and memory allocation parameters. 

Ref:
http://docs.neo4j.org/chunked/milestone/configuration-caches.html
http://docs.neo4j.org/chunked/stable/configuration-io-examples.html
http://docs.neo4j.org/chunked/stable/configuration-jvm.html

Principally if you make enough memory available to Neo4j, manually set 
memory map sizes and opt for strong caching you should get the equivalent 
of an in-memory graph after one traversal over the whole graph. With strong 
caching Neo4j doesn't ever kick anything out of cash once loaded.

Cheers,
Mah

-- 
You received this message because you are subscribed to the Google Groups 
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to