I’m continuing to post Gluecon session descriptions (check out the agenda). Here’s Nathan Marz:
Exporting data from Hadoop with ElephantDB
ElephantDB is a database that makes it easy to export data out of Hadoop to serve in your applications. Both the exporting and serving of a dataset is done in a completely distributed way. ElephantDB emphasizes ease of use and stability — it just works. Unlike alternatives like HBase, ElephantDB is easy to configure, requires almost no maintenance, and is easy to deploy. BackType uses ElephantDB to serve hundreds of gigabytes of views off of a 30TB dataset.
Since ElephantDB can only be written to in batch, and doesn’t support random writes, it is extremely simple. This leads to ElephantDB being rock-solid in production, since there’s almost no moving parts.
In this session, you’ll learn how to use ElephantDB and how to deploy/operate it. You’ll learn how the exporting process works and how to configure an ElephantDB cluster to serve that data. Additionally, you’ll learn how to use ElephantDB’s automated deploys to provision and launch an ElephantDB cluster with just the click of a button. We’ll see how to do incremental updates to an ElephantDB domain, and we’ll see how to use ElephantDB as an indexed key/value file format on top of Hadoop.