InfiniteGraph User Forums on Google Groups

In order to facilitate more interactive, real-time discussions among our visitors, we’ve created a new Google Group that will host your topics, threads and discussions. Please sign up today so you can ask the experts for advice, or help new users by sharing what you’ve already learned!

Google Groups

Visit InfiniteGraph’s Google Group

Thanks everyone, and we look forward to hearing from you!
The InfiniteGraph Team


  1. aris setyawan
    Posted June 30, 2010 at 9:17 pm | Permalink

    As I search in months, no distributed (sharded) graph database exist, until InfinitGraph sends me an email. I have some questions directly to the graph implementation. I will ask you one and in the following posts. How InfinitGraph partition big graph and how to merges those?

  2. aris setyawan
    Posted June 30, 2010 at 9:18 pm | Permalink

    What is the back-end storage use by InfinitGraph?

    • Arvind G
      Posted July 1, 2010 at 2:04 pm | Permalink

      InfiniteGraph uses Objectivity/DB for the back end storage. This allows in memory representation of graph objects to be stored on disk, and can be accessed via any OS and hardware platform that we support. It also allows for our distributed architecture.

  3. aris setyawan
    Posted June 30, 2010 at 9:19 pm | Permalink

    How I can traversal to my Graph in InfinitDB API? How flexible is the API? Can U give me an example?

    • Arvind G
      Posted July 1, 2010 at 2:58 pm | Permalink

      The beta download has a basic sample called GraphAPISample. Although the usage there is pretty basic, it can be extended- a custom “Guide” can be implemented to guide the navigation, custom result and navigation “Qualifier”s can be implemented to validate and terminate paths in your navigation. The navigation results can be passed to a a user implementation of NavigationResultHandler, where you can do print the paths, etc.

  4. aris setyawan
    Posted June 30, 2010 at 9:20 pm | Permalink

    What kinds of index provided in InfinitDB? Can U give me an example in how it’s created and can speed up my query?

    • Arvind G
      Posted July 1, 2010 at 3:08 pm | Permalink

      There are at least three kinds of indexing available in InfiniteGraph. The first is to use the nameVertex() function, so you can set retrieve vertices by name. This has limited scalability.

      Two others are in the indexing package, namely GraphIndex and LuceneIndex.

      The second way to index objects is to use the GraphIndex class which provides automatic indexing for objects and their fields via a number of static methods.

      The third is to use the LuceneIndex class (which has a dependency on Lucene jar file available at

      There are some code snippets available in the javadocs (available in the download) for the classes listed above. If you still need more help, let us know by email, and I can create some more fully functional samples.

  5. aris setyawan
    Posted June 30, 2010 at 9:22 pm | Permalink

    In traversal, can I limit the number of row we will retrieve? (some kind like LIMIT i, OFFSET j). I need it to work with huge size graph.

    • Arvind G
      Posted July 1, 2010 at 3:30 pm | Permalink

      Yes you can. You can limit the number of paths you examine. Within your NavigationResultHandler implementation you can have a integer attribute for number of results handled thus far. Increment this integer every time you handle a result. Once the number reaches the limit you want, call Navigator.stop() to stop navigating.

  6. aris setyawan
    Posted June 30, 2010 at 9:23 pm | Permalink

    Although I save my node and edge in form of graph, can I retrieve it (eg, one kind of node) in the form of table? With sorting options?

    • Arvind G
      Posted July 1, 2010 at 3:32 pm | Permalink

      Yes, you might want to use our GraphIndex or LuceneIndex classes in the indexing package. The index lookups return a java.util.List, which can be sorted any way you want.

  7. CordiS
    Posted July 20, 2010 at 6:29 am | Permalink

    Are you going to develop some kind of REST API or even python bindings?

    • Posted August 11, 2010 at 1:42 pm | Permalink

      While we are evaluating several new features and bindings for subsequent versions, the focus for the next release will be on improving transactional concurrency and graph traversal.

Post a Comment

Required fields are marked *

%d bloggers like this: