Streaming graph data partitioning issue

[thamindumk]

When a graph stream into the JasmineGraph system, the graph should be partitioned into different partitions. the system will run three partition algorithm according to the node id. the system requires a node id to run the partition algorithm. When the system uses the hash partitioning algorithm, it wants integer id values to partition the data. The issue is, the id can be a uuid. in that case system wants to have a proper partitioning technique to generate same partition id for unique uuid or other unique id value.

as an example:
uuid-1 : 550e8400-e29b-41d4-a716-446655440000 -> partition id : 3
uuid-2 : 361ceb3f-6082-4abc-b1d3-44e61df1fed1 -> partition id : 2

and again if the system receives the uuid-1, it must generate the same partition id, 3. Or otherwise the node is stored in multiple workers and it will be a confusion when querying the data.

and another case is, if the received data doesn't have a property called "id", it will cause another issue