Reordering Strategy for Blocking Optimization in Sparse Linear Solvers
From MaRDI portal
Publication:5346752
DOI10.1137/16M1062454zbMath1366.65047OpenAlexW4386017381MaRDI QIDQ5346752
Gregoire Pichon, Mathieu Faverge, Pierre Ramet, Jean Roman
Publication date: 29 May 2017
Published in: SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1137/16m1062454
algorithmnested dissectionsparse matrix orderingheterogeneous architecturessparse direct solverblock-symbolic factorizationsparse block linear solver
Computational methods for sparse matrices (65F50) Direct numerical methods for linear systems and matrix inversion (65F05)
Related Items
Fast Implementation of the Traveling-Salesman-Problem Method for Reordering Columns within Supernodes, Hierarchical Orthogonal Factorization: Sparse Square Matrices
Uses Software
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Algorithmique et calculs de complexité pour un solveur de type dissections emboîtées. (Algorithmic study and complexity bounds for a nested dissection solver)
- On the Minimum FLOPs Problem in the Sparse Cholesky Factorization
- The university of Florida sparse matrix collection
- The Role of Elimination Trees in Sparse Factorization
- A Separator Theorem for Planar Graphs
- On Finding Supernodes for Sparse Matrix Computations
- An Analysis of Several Heuristics for the Traveling Salesman Problem
- Algorithmic Aspects of Vertex Elimination on Directed Graphs
- On the Application of the Minimum Degree Algorithm to Finite Element Systems
- A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs
- A set of level 3 basic linear algebra subprograms
- Error Detecting and Error Correcting Codes
- An Approximate Minimum Degree Ordering Algorithm
- Design of a Multicore Sparse Cholesky Factorization Using DAGs
- Implementing Multifrontal Sparse Solvers for Multicore Architectures with Sequential Task Flow Runtime Systems
- Nested Dissection of a Regular Finite Element Mesh
- PaStiX: A high-performance parallel direct solver for sparse symmetric positive definite systems
