F1-3 MPI-OpenMP Hybrid Parallelization of Iterative Solvers for the FrontISTR

スポンサーリンク

概要

• 論文の詳細を見る

• High-end computers such as T2K or the K computer requires hybrid parallel programming model because of their hierarchical hardware architecture. Conjugate gradient solver in a MPI parallel structural analysis application FrontISTR that invokes HEC-MW is parallelized in a MPI-OpenMP hybrid way. In the current implementation, 1.5 - 3.0 x speedup in SpMV operations and 1.5 x speedup in actual preconditioned CG solver have been achieved.

• 2010-09-23

著者

• OKUDA Hiroshi

  Research and Education Center for Materials Science, Nara Institute of Science and Technology

• SATO Yohei

  Institute of Multidisciplinary Research for Advanced Materials Tohoku University

• Okuda Hiroshi

  Research Into Artifacts Center For Engineering University Of Tokyo

• WATANABE Tatsuru

  Graduate School of Engineering, University of Tokyo

• Watanabe Tatsuru

  Graduate School Of Engineering University Of Tokyo

• Sato Yohei

  Institute Of Industrial Science University Of Tokyo

関連論文

• バングラデシュにおける洪水のモニタリング:通常の洪水と大規模な被害をもたらす洪水
• Oscillatory Resonant X-Ray Magnetic Bragg-Peak Signs and Gd Moments Induced by the Fe/Gd Interlayer Exchange Interaction(Magnetism)
• High energy-resolution electron energy-loss spectroscopy study of the electric structure of double-walled carbon nanotubes
• Finite element analysis of crustal deformation in the Ou Backbone Range, northeastern Japan, with non-linear visco-elasticity and plasticity : effects of non-uniform thermal structure
• Shape Optimization Using Adjoint Variable Method for Reducing Drag
• Feasibility Study of GPU-Accelerated Conjugate Gradient Solver
• F1-3 MPI-OpenMP Hybrid Parallelization of Iterative Solvers for the FrontISTR
• 24aTG-5 TEM/EELS studies of polyynes inside SWCNT
• 21aEC-8 TEM/EELS studies of coronenes inside single-walled carbon nanotube
• Convergence Acceleration of Parallel CG-FEM with Controlled Domain Decomposition for Singularity Problems

スポンサーリンク