From what I understand HBase is a little more feature complete. It has row locking built in for instance. Cassandra really can't do that by the very nature of

From what I understand HBase is a little more feature complete. It has row locking built in for instance. Cassandra really can't do that by the very nature of how it is designed(fully peer to peer, no zookeeper). On the other hand Cassandra is really easy to get going, just download a build and double click. Hbase requires zookeeper and hadoop\hdfs before you can get it running as such it is more complicated and the config is a bit more squirrelly. It really is MySQL vs Postgresql all over again.