Thanks - that's great info. I'll try more granular batches. On Sat, Feb 20, 2010 at 4:26 PM, Raul Raja Martinez <[email protected]>wrote:
> I think the limitation is more related to the size of the transaction. > Less transactions = faster inserts = more memory > Just a guess though, not sure what your code looks like. > We have been able to import millions of nodes with small heaps committing > in > small batches. > > 2010/2/20 Jason Boorn <[email protected]> > > > Hi - > > > > I wanted to get a sense of how heap memory is being used in the database. > > I > > have a small dataset where nodes store arrays of strings, and I can't > seem > > to load in more than about 50K nodes (each node has about 100 bytes in > its > > string array). I get out of memory errors. > > > > My question specifically is the following: is the dataset I work with > > explicitly limited by the heap size available to the JVM? In other > words, > > if I have a data set where each node's properties amount to 1K, and I > have > > a > > 1GB heap, does this mean my database can only contain 1MM nodes? > > > > Thanks > > _______________________________________________ > > Neo mailing list > > [email protected] > > https://lists.neo4j.org/mailman/listinfo/user > > > > > > -- > Raul Raja > _______________________________________________ > Neo mailing list > [email protected] > https://lists.neo4j.org/mailman/listinfo/user > _______________________________________________ Neo mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
