There's an O'Reilly book in the works that I think has a preview available.
The flexibility of Accumulo is that it allows you to create your own serialization mechanisms to store/retrieve data. This gives you the ability to structure data in such a way that is efficient for your use cases. The problem, as you stumbled upon, is that this can be rather overwhelming.
There are lots of examples that you can look at to help provide some examples (e.g. Pig[1], Jaccson[2], OpenTSDB[3], or just look at Github[4]). I'm not aware of any library that's general-purpose enough for universal use.
[1] http://svn.apache.org/repos/asf/pig/trunk/src/org/apache/pig/backend/hadoop/accumulo/AccumuloStorage.java
[2] https://github.com/acordova/jaccson [3] https://github.com/ericnewton/accumulo-opentsdb [4] https://github.com/search?q=accumulo&type=Repositories On 3/2/14, 11:55 PM, Kevin Cho wrote:
Currently, I'm trying to learn the basics to get through interview. So far, it seems pretty easy to use. I was able to install Accumulo in cluster mode and do some basic admin and table cruds. However... I'm really puzzled on how to architect DAO around Accumulo. Do I create DAO per column family? Is there a another library like hibernate where it would automatically deserialize to Java objects? Thanks again! If you have recommended Accumulo book then please let me know as well.
