Thanks to gil for the review!
The glusterfs-hadoop package is now available in F20+. It provides an
implementation of Hadoop's distributed filesystem 1.x and 2.x interfaces
(
http://wiki.apache.org/hadoop/HCFS) that work with GlusterFS.
The feature provided is the ability to run any computational workloads
(MapReduce, HBase, Hive, etc etc) on top of data stored in Gluster.
The project's homepage is
https://forge.gluster.org/hadoop
Best,
matt