Massively Parallel Heuristic Search with MapReduce
Heuristic search can be applied to a wide range of optimization problems, e.g. to multiple sequence alignment or shortest path problems. In the MR-Search project, we are developing scalable search algorithms for massively parallel supercomputers.
MR-Search is a framework for massively parallel breadth-first and heuristic search based on the MapReduce paradigm. It can run on shared-memory machines using OpenMP and on distributed memory machines using MPI.
Thorsten Schütt, Alexander Reinefeld, Robert Döbbelin