Shared-memory implementation of the Karp-Sipser kernelization process
Résumé
We investigate the parallelization of the Karp-Sipser kernelization technique, which constitutes the central part of the well-known Karp-Sipser heuristic for the maximum cardinality matching problem. The technique reduces a given problem instance to a smaller but equivalent one, by repeated applications of two operations: vertex removal, and merging two vertices. The operation of merging two vertices poses the principal challenge in parallelizing the technique. We describe an algorithm that minimizes the need for synchronization and present an efficient shared-memory parallel implementation of the kernelization technique for bipartite graphs. Using extensive experiments on a variety of multicore CPUs, we show that our implementation scales well up to 32 cores on one socket.
Origine | Fichiers produits par l'(les) auteur(s) |
---|