Once we have a dummy partitioner (ticketed in #3204) we unblock further steps in the pipeline. Eventually we want to replace the dummy partitioner with a proper partitioning approach: Inertial Flow.

On Balanced Separators in Road Networks
Aaron Schild and Christian Sommer
SEA 2015 - 14th International Symposium on Experimental Algorithms (pp. 286-297)
doi: 10.1007/978-3-319-20086

See: http://sommer.jp/roadseparator.htm

We already successfully prototyped a internal routing engine with that - some hints are over at

http://daniel-j-h.github.io/post/selection-algorithms-for-partitioning/

our multiple candidate cuts and parallelization should be doable in osrm, too.

Tasks:

• Dinic's Max Flow / Min Cut
• Inertial Flow