SuperLU_DIST: A scalable distributed-memory sparse direct solver for unsymmetric linear systems

By Xiaoye S. Li, James Weldon Demmel. TOMS, 2003

736 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/ed16cedb73de948c9ca4ac4d080250b5be86a236

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiaoye Sherry Li
2. Anton Evgrafov
3. Klaus Gärtner
4. Kurt Maute
5. G. Pingen
6. Olaf Schenk

In Computational Theory and Mathematics

1. Kurt Maute
2. G. Pingen
3. Klaus Gärtner
4. Olaf Schenk
5. Xiaoye Sherry Li
6. Anton Evgrafov
7. David J. Gardner
8. Daniel R. Reynolds

In Hardware and Architecture

1. Klaus Gärtner
2. Olaf Schenk
3. Xiaoye Sherry Li
4. Anton Evgrafov
5. Kurt Maute
6. G. Pingen
7. Rolf H. Krause
8. Thomas Dickopf

Citations in the corpus

• An overview of SuperLU: Algorithms, implementation, and user interface (2003, methodology): “We observed 20% to 40% uniprocessor performance improvement [Li and Demmel 2003].”
• Enabling New Flexibility in the SUNDIALS Suite of Nonlinear and Differential/Algebraic Equation Solvers (2020, background): “The current set of SUNDIALS-provided SUNLinearSolvers includes direct solvers for dense (native, LAPACK [2], MAGMA [73], and oneMKL [46]), banded (native and LAPACK), and sparse (KLU [23, 24], SuperLU_MT [26], SuperLU_Dist [54], and cuSOLVER [58]) matrix-based linear systems, as well as an ensemble ”
• Topology optimization of flow domains using the lattice Boltzmann method (2007, methodology): “For practical reasons, we apply a parallel direct sparse solver to this problem, namely the parallel version of the SuperLU solver (see Li and Demmel 2003 ), acknowledging that greater parallel scalability could be obtained with iterative solvers.”
• Stable Generalized Finite Element Method (SGFEM) (2011, background): “Problem 2a: The linear system is solved by a sparse matrix direct solver superLU [29] on a single processor.”
• Solvers for the Cardiac Bidomain Equations (2007, background): “, 2004a), this problem could be possibly circumvented by employing a parallel direct solver (Li and Demmel, 2003) at the coarsest grid, the additional communication burden imposed by a parallel direct method, however, has to be dealt with by using a low latency interconnect like InfiniBand.”
• ON FAST FACTORIZATION PIVOTING METHODS FOR SPARSE SYMMETRI C INDEFINITE SYSTEMS (2006, methodology): “In [24] the pivoting approach of SUPERLU is described for the factorization of sparse nonsymmetric systems.”