Apache HAMA is a distributed computing framework based on Bulk Synchronous Parallel algorithms. It provides high performance computing for performance-intensive, scientific, and iterative algorithms such as Matrix, Graph, and Machine Learning.
Apache Giraph is a framework for iterative processing of semi-structured graphs. It is inspired from Google's Pregel, which is also a graph processing framework. Giraph is also based on a Bulk Synchronous Parallel model of distributed computing.
For more details on Giraph, you can refer to the official website at http://giraph.apache.org/.
Initially, Giraph was used with the MapReduce framework for Hadoop 1.x. There were a few concerns, such as: