ویراساینس.
سند علمی

MUMPS : A General Purpose Distributed Memory Sparse Solver

Patrick Amestoy, Iain S. Duff, Jean-Yves L'Excellent, Jacko Koster

چکیده

MUMPS is a software package for the multifrontal solution of large sparse linear systems on distributed memory computers. The matrices can be symmetric positive definite, general symmetric, or unsymmetric, and possibly rank deficient. MUMPS exploits parallelism coming from the sparsity in the matrix and parallelism available for dense matrices. Additionally, large computational tasks are divided into smaller subtasks to enhance parallelism. MUMPS uses a distributed dynamic scheduling technique that allows numerical pivoting and the migration of computational tasks to lightly loaded processors. Asynchronous communication is used to overlap communication with computation. In this paper, we report on recently integrated features and illustrate the present performance of the solver.

اسناد مشابه

  • Performance and tuning of two distributed memory sparse solvers

    سند علمی۲۰۰۱PPSCPatrick Amestoy, Iain S. Duff, Jean-Yves L'Excellent, Xiaoye S. Li

    We examine the performance of two codes using direct methods to solve large sparse linear equations on distributed memory computers. The code SuperLU is a right-looking supernodal code, written in C, while MUMPS is written in Fortran 90 and uses a multifrontal technique. Both cod

  • A distributed-memory hierarchical solver for general sparse linear systems

    سند علمی۲۰۱۷CoRRChao Chen, Hadi Pouransari, Sivasankaran Rajamanickam, Erik G. Boman

    We present a parallel hierarchical solver for general sparse linear systems on distributed-memory machines. For large-scale problems, this fully algebraic algorithm is faster and more memory-efficient than sparse direct solvers because it exploits the low-rank structure of fill-i

  • P-sparslib: a Portable Library of Distributed Memory Sparse Iterative Solvers

    سند علمی۱۹۹۵Yousef Saad, Andrei V. Malevsky

    Domain Decomposition techniques constitute an important class of methods which are especially appropriate in a parallel computing environment. However, few general purpose computational codes based on these techniques have been developed so far. In this paper, we propose one such

  • Kernels for Distributed Memory Sparse Iterative Solvers ?

    سند علمی۱۹۹۵Yousef Saad, Andrei V. Malevsky

    Domain Decomposition techniques constitute an important class of methods especially appropriate in a parallel computing environment, but only a few general purpose codes based on these techniques have been developed so far. In this work, we attempt to develop not only algorithms

  • Task Scheduling in an Asynchronous Distributed Memory Multifrontal Solver

    سند علمی۲۰۰۴SIAM J. Matrix Analysis ApplicationsPatrick Amestoy, Iain S. Duff, Christof Vömel

    We describe the improvements to the task scheduling for MUMPS, an asynchronous distributed memory direct solver for sparse linear systems. In the new approach, we determine, during the analysis of the matrix, candidate processes for the tasks that will be dynamically scheduled du

  • Trading Replication for Communication in Parallel Distributed-Memory Dense Solvers

    سند علمی۲۰۰۲Parallel Processing LettersDror Irony, Sivan Toledo

    We present new communication-efficient parallel dense linear solvers: a solver for triangular linear systems with multiple right-hand sides and an LU factorization algorithm. These solvers are highly parallel and they perform a factor of 0.4P1/6 less communication than existing a