Then it seems that heroku is quite a good fit for you. Register for their beta program (b...@heroku.com) then you can add the neo4j server to your app. The servers are not in germany as peter said but co-located with heroku at AWS US-EAST.
With the heroku add-on you can just zip your embedded graph-db data directory (just the files) and upload it to the server. With our current ruby script extension you can write a server-side rack (sinatra|rails) app that provides the domain level REST endpoints (which have a much better granularity, transactional behaviour and performance than the low level REST API). (see here: http://wiki.neo4j.org/content/Neo4j_Heroku_Addon, and http://blog.neo4j.org/2011/08/heroku-neo4j-add-on-available-in.html) As Peter said for querying you can also look into cypher and gremlin. There are also some ready-made AWS images for Neo4j server provided by OpenCredo. Cheers Michael Am 18.09.2011 um 16:10 schrieb Peter Neubauer: > John, > great to hear things worked out for you! Let me answer what I can inline ... > > On Sun, Sep 18, 2011 at 3:38 PM, John Doran <john.do...@hotmail.com> wrote: >> The last project I did with Neo4J was a success(well I see it as one), I >> used the embedded graph db with GWT and some cool other stuff. The graph >> consosted of a spatial layer and a normal layer. So my plan is to move to >> the server and do some mobile apps(iPhone then Android) that can hit the >> server, it will be read only as all the data will be static and populated >> initially by myself. >> > Anything you are missing from Neo4j Spatial btw? > >> So here are my questions - >> Is there any way I can leverage my current graph db set-up with the RESTful >> server? Some way to do a data dump maybe? >> >> Another though I have had was, I have this graph db set-up. maybe I can just >> use it and set up some servlets/simple J2EE server that will let me hit it >> for request, and actually just use the embedded graph-db, just throwing the >> responses back in JSON or XML. > Yes, just look at the Neo4j Server installation, > http://docs.neo4j.org/chunked/snapshot/server.html and the REST docs, > http://docs.neo4j.org/chunked/snapshot/rest-api.html > >> >> Also, I don't want the trouble of looking after a server on a machine, is it >> possible to get a service provider to look after such a set-up? >> a) Using the J2EE and embedded graph setup. >> b) The RESTful Neo4J server. > We have a Heroku beta test running, see > http://addons.heroku.com/neo4j, in the backend provisioning Neo4j > Server instances on a fabric maintained in Germany. Also, there is > work in CloudFoundry going on to have Neo4j being part of it, as is on > Microsoft Azure. > > Would some of these be ok for you? > >> >> How much of a difference is there between working with the server? What's >> the learning curve involved? My only experience with REST is RoR. > Well, should not be that hard, you can do graph operations via the > very convenient client bindings for Ruby by Maxz De Marzi, > https://github.com/maxdemarzi/neography . > > For querying and doing more complex stuff on the server in order to > reduce chattiness, look at the Cypher and Gremlin plugins, > http://docs.neo4j.org/chunked/snapshot/cypher-plugin.html and > http://docs.neo4j.org/chunked/snapshot/gremlin-plugin.html > > Let us know how things go for you! > > /peter > _______________________________________________ > 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
