PT-Scotch: A tool for efficient parallel graph ordering

Cédric Chevalier (1) , François Pellegrini (1, 2)

1 SCALAPPLIX - Algorithms and high performance computing for grand challenge applications
2 LaBRI - Laboratoire Bordelais de Recherche en Informatique

Graph partitioning is an ubiquitous technique which has applications in many fields of computer science and engineering. Because there always exists large problem graphs which cannot fit in the memory of sequential computers and cost too much to partition, parallel graph partitioning tools have been developed. Graph ordering is the first target application of the PT-Scotch (“Parallel Threaded Scotch”) software, a parallel extension of the sequential Scotch graph partitioning and ordering tool that we are currently developing within the ScAlApplix project. We outline in this extended abstract the algorithms which we have implemented in PT-Scotch to parallelize the Nested Dissection ordering method.

Mots clés: Parallel computing; sparse matrix ordering

Domaines: Calcul parallèle, distribué et partagé [cs.DC]

Cédric Chevalier, François Pellegrini. PT-Scotch: A tool for efficient parallel graph ordering. 4th International Workshop on Parallel Matrix Algorithms and Applications (PMAA'06), Sep 2006, Rennes, France. ⟨hal-00410427⟩