dc.contributor.author | Hu, Yu Charlie | |
dc.contributor.author | Teng, Shang-Hua | |
dc.contributor.author | Johnsson, S. Lennart | |
dc.date.accessioned | 2016-02-25T18:25:28Z | |
dc.date.issued | 1996 | |
dc.identifier.citation | Hu, Yu Charlie, Shang-Hua Teng, and S. Lennart Johnsson. 1996. A Data-Parallel Implementation of the Geometric Partitioning Algorithm. Harvard Computer Science Group Technical Report TR-15-96. | en_US |
dc.identifier.uri | http://nrs.harvard.edu/urn-3:HUL.InstRepos:25620495 | |
dc.description.abstract | We present a data-parallel, High Performance Fortran (HPF) implementation of the geometric partitioning algorithm. The geometric partitioning algorithm has provably good partitioning quality. To our knowledge, our implementation is the first data-parallel implementation of the algorithm. Our data-parallel formulation makes extensive use of segmented prefix sums and parallel selections, and provide a data-parallel procedure for geometric sampling. Experiments in partitioning particles for load-balance and data interactions as required in hierarchical N-body algorithms and iterative algorithms for the solution of equilibrium equations on unstructured meshes by the infinite element method have shown that the geometric partitioning algorithm has an efficient data-parallel formulation. Moreover, the quality of the generated partitions is competitive with that offered by the spectral bisection technique and better than the quality offered by other partitioning heuristics. | en_US |
dc.description.sponsorship | Engineering and Applied Sciences | en_US |
dc.language.iso | en_US | en_US |
dash.license | LAA | |
dc.title | A Data-Parallel Implementation of the Geometric Partitioning Algorithm | en_US |
dc.type | Research Paper or Report | en_US |
dc.description.version | Version of Record | en_US |
dc.date.available | 2016-02-25T18:25:28Z | |