...
The internals of the Graph module implemented on top of BSP framework, are pretty simple. Input data partitioning will be done at BSP framework level. Each GraphJobRunner processors just reads assigned splits, and loads parsed vertices into memory-based vertices store storage at loadVertices() method. (Currently disk-based vertices store and sequential vertex processing are not perfect)If you want to learn details and internals about Graph job, Please see also Design of Graph Module.
Create your own Partitioner
...