Thursday, December 2, 2010

Neo4j 1.2 Milestone 5 - Reference Manual and HA!

Peter Neubauer

Hi there,
in this release, the community and the Neo4j team have been working hard to bring you some really long-term value in the Neo4j project, mainly Documentation and the first glimpse of the High Availability component.

Thanks to all individuals reporting bugs, feedback and patches, a couple of them got into this release, like for instance:
  • Minor Neo4j server script enhancements from Mao Pu
  • REST API feedback form Rick Bullotta, Max De Marci and many others
  • Much feedback on visualization options for graphs from the community


Neo4j now has the beginnings of a one-stop reference manual as both online HTML and PDF, giving you reference documentation on using Neo4j and the related components. We hope this will make it much easier to find relevant information, getting started and look up things that previously have been spread over the wiki, blog posts and other different online sources. This documentation is available online at .
This is work in progress, but please get back to us with feedback on what you want to see more of, things that might be missing and other comments!

Also, this documentation will be versioned, so that for each Neo4j release there will be a corresponding version of the Neo4j Reference Manual. Eventually, we hope to evolve this into a book on Neo4j. If you want to contribute to the effort, let us know!

High Availability

With this milestone, we are proud to for the first time release a version of the Highly Available Neo4j cluster to the community. This component turns Neo4j into a system with a Write-Master and a number of Read-Slaves. Upon mutating operations, at least the master and one slave are committing the transaction before propagating out the changes to the other slaves. In case of the write-master failing, the remaining slaves will automatically elect a new master. Also, the whole cluster infrastructure is transparent to the code that uses the Neo4j API, so that all slaves and masters can be written to and read from transparently. Your code does not need to be aware of the type of HA setup it is running on. Simply do

GraphDatabaseService db = new HighlyAvailableGraphDatabase( path, config );
and off you go as usual!
More on that to come as part of the documentation, but feel free to get it going and give it a spin.

As always, to get the release, download it from or update your pom.xml to


May the source be with you!


Friday, November 5, 2010

Neo4j 1.2 Milestone 3 - Habit Forming

Andreas Kollegger

The first time is a promise. The second, a confirmation. But it's not until the third time you follow through on cleaning your room that you can claim it as a habit.

We're feeling good about having a milestone release habit. Regular releases keep us mindful of always working towards delivering something useful. Milestone 3 has been quality-assured to minor API changes, bug fixes and some repackaging.

Get it from the download site. And read about the changes here.

Oh, and a reminder for anyone in the glorious Øresund region: look for Neo4j team members at Øredev next week.




Tuesday, October 26, 2010

Neo4j at Öredev and Oredev Open, Nov 8-12, Malmö, Sweden

Peter Neubauer

Hi all,
the Neo4j team is very excited to announce some Neo4j Team members being present at this years Öredev. Jim Webber is talking about REST in Practice, and Emil Eifrem is introducing the new Spring Data && NoSQLcomponents, with Neo4j being the first fully supported backend.

Also, there is the Öredev Open event on November 11. where MongoDB and Neo4j will contribute to the NoSQL track. Come, join and share - Malmö is always worth a visit in November :)


Your Neo4j team.

Thursday, October 21, 2010

Neo4j 1.2.M02 is out

In following our new bi-weekly release schedule, we today have released version 1.2.M02 of Neo4j. It's been a productive couple of weeks, and the biggest change for our users is the new integrated index API, with Lucene as the default implementation.

Update your maven pom dependencies to:


More links:

Download here
Read more here
Get started here

Happy hacking


Tuesday, October 12, 2010

Neo4j 1.2 Milestone 1 released

Peter Neubauer

The Neo4j team is pleased to announce the first milestone towards the 1.2 release of Neo4j -- 1.2.M01 being the version number. Our current focus is on usability, which means improving the overall experience of getting, installing, and using Neo4j. We're starting with:
  • a bi-weekly release schedule, for quick introduction of new features
  • deploying artifacts to maven central, for hassle-free distribution
  • new ways of gathering feedback, so we can figure out what works
Depend on it here --

Tuesday, August 17, 2010

Neo4j visualization with Gephi

During the summer Martin Škurla has developed support for the Neo4j graph database in the Gephi visualization and exploration platform. This Google Summer of Code project is now approaching its finish.

The basic idea of the project is twofold:

  1. Users of Neo4j get better visualization support
  2. Users of Gephi can work with larger graphs

The graph visualization can give you output like this:

Martin just published an article summarizing the project, head over and read it! For the future development of the Neo4j plugin, there's a survey where everyone has the chance to give feedback on what features they want to use.

To try it out, download NetBeans 6.9 and use it to build and execute the source code (Gephi depends on the NetBeans platform version 6.9). To check out the source, make sure to have Bazaar installed, and then do:

bzr checkout lp:~bujacik/gephi/support-for-neo4j

Finally, you're welcome to discuss the plugin on the Neo4j mailing list and in the Gephi forums.

Update: See the comments for information on configuring NetBeans to build Gephi!

Thursday, June 3, 2010

NOSQL summer in Malmö!

Hi all,
the summer is finally here, and with it soooo much spare time. Why not meet up with some peeps and discuss some of the cool new paradigms of database technology popping up in the NOSQL space?

Tim Anglade has been taking up the flag and is pulling off the NOSQL summer. If you are up for some cool discussions, join one of the places! NeoTechnology is providing space and beer for a first meetup 16th of June in Malmö. We are starting off the the Amazon Dynamo paper, a classic in new data thinking. If you are in Sweden or Denmark, ping us on the list and come along!

Thursday, April 15, 2010

Neo4j Standalone REST server 0.8 available

The Neo4j team is proud to announce the availability of a REST wrapper that can be installed both on Windows, Linux and Mac OS X as a standalone server. Please follow the Getting Started guide on the Neo4j wiki where you'll find download links, the Matrix-REST example is available over here.

This makes the interaction with the Neo4j graph database from PHP, .NET and other languages like Perl a lot easier. Let us know how things work out via the Neo4j mailing list!

Thursday, March 25, 2010

Google Summer of Code Projects with Neo4j

Great news for all students wanting to spend the summer hacking on Neo4j and getting paid for it! The hackers behind Neo4j have teamed up with a few mentoring organizations on some interesting Summer of Code projects!

So far there are two proposed projects involving Neo4j, but more projects could be added in the next couple of days. You could of course also come up with a cool project on your own and propose it to one of the mentoring organizations.

The first project we have proposed is a collaboration with the uDIG project. This project is part of our greater effort of making Neo4j capable of being a fully featured spatial database. The sweet spots are complex spatial algorithms such as routing, spatial analytics and N-dimensional indexing. This summer of code project will aim at implementing the OpenGIS and GeoTools standard interfaces. This will make Neo4j a drop-in alternative to other GIS backends, such as PostGIS, MySQL Spatial and Oracle Spatial.

The second project we have proposed is a collaboration with the Gephi project. What we aim for here is to make Gephi able to visualize graphs straight out of a Neo4j graph (not just exporting data from Neo4j and importing it in Gephi). This will enable Gephi to work with larger graphs than today, which is a direct benefit for Gephi. It will also enable Gephi to be used as a visualization tool for Neo4j based applications, for introspection, data debugging and graphical reporting.

This will be a truly great summer for Neo4j and the lucky students who get selected for these projects. There's a lot of interest in both GIS and visualization in the Neo4j community, so we look forward to great discussions around these projects! Mark March 29 in your calendar and submit your proposal early!

Friday, March 12, 2010

Update on Neo4j Ruby bindings

During 2010 there's been two releases of the Neo4j.rb JRuby bindings for the Neo4j graph database so far. Time to catch up with what's new!

Version 0.4.0 of Neo4j.rb came with improved traversal performance, more options on how to use relationships, for instance relationships can now be indexed. Version 0.4.1 gave us migrations and access to a batch inserter for big import-once data volumes.

In the last days of 2009, neo4jr-simple was first released. It's a simple ready to go wrapper for Neo4j and currently in version 0.2.1. Make sure to check out the neo4jr-social example application as well!

Great thanks to Andreas Ronge, Matthew Deiters and all the other contributors for the awesome stuff!