Hi there, I wanted to look into more efficient ways to send back big responses, but had not time. Basically, throwing in more memory is what you have to do right now, or get Cypher or Gremlin to return more effective representations like only the properties you need, which cuts down the amount of data sent over the wire dramatically, plus chunking, see http://docs.neo4j.org/chunked/snapshot/query-skip.html
Would that be a possibility Cheers, /peter neubauer GTalk: neubauer.peter Skype peter.neubauer Phone +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter http://twitter.com/peterneubauer http://www.neo4j.org - NOSQL for the Enterprise. http://startupbootcamp.org/ - Öresund - Innovation happens HERE. On Fri, Nov 11, 2011 at 9:35 AM, Kevin Versfeld <kevin.versf...@gmail.com> wrote: > hi Michael/Peter > > Peter: yes, that about sums it up. Webadmin console is fine, but REST not so > much. > > Michael: my final set is a little over 100K nodes, and while I am still > playing around as part of a technical proof of concept, I think I will need > this data more or less in its entirety. > > I pumped up the memory settings in the neo configs, and increased my rest > timeout, and managed to get a successful response (at which point my > consuming code died, I think at the deserializing point :) ), so then I > tried some manual chunking (using paged results), which worked pretty well, > so I think that is the route I'll need to take. > Also noteworthy, I have been using Tatham and Romiko's Neo4jClient stuff so > far, and was just chatting via email to Romiko earlier, and he said that > their client does support chunking already (at least in parts?) - so I will > get latest and see what I can see. > > I think my main concern or question left is around the memory usage of my > java.exe process: although everything now executes, its still not terribly > fast (although considering its over REST, and is doing a lot of > deserializing etc, I think its acceptable). Most importantly though, I see > my java.exe process still consumes a lot of memory (half an hour later, and > its still sitting at a gig...) - should I be worried about this? > > -- > View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/Gremlin-how-to-flatten-a-tree-and-sort-tp3480586p3499114.html > Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. > _______________________________________________ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user > _______________________________________________ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
