BerkeleyDB is suitable for smaller data sets up to 10 million vertices with ACID guarantees. It can be used for proofs of concept or testing in developer environments. On the other hand, HBase does not provide ACID guarantees, but is capable of scaling for large graphs. In addition, HBase inherently provides high availability and generally has higher expected response times while creating types and entities than does BerkeleyDB. For further details, see http://s3.thinkaurelius.com/docs/titan/0.5.4/bdb.html and http://s3.thinkaurelius.com/docs/titan/0.5.4/hbase.html.