Towards efficient and scalable graph partitioning methods
Résumé
The realization of efficient parallel graph partitioners requires the parallelization of the multi-level framework which is commonly used in sequential partitioners to improve quality and speed. While parallel matching algorithms are now efficient and un-biased enough to yield coarsened graphs of good quality, the local optimization algorithms used in the refinement step during uncoarsening are still an issue. This talk will address this problem and present scalable parallel diffusive methods which can advantageously replace classical Fiduccia-Mattheyses-like algorithms for this purpose.