SparseLU, A Novel Algorithm and Math Library for Sparse LU Factorization

Conference · Tue Nov 01 00:00:00 EDT 2022

DOI:https://doi.org/10.1109/IA356718.2022.00010· OSTI ID:2000278

Valero Lara, Pedro ^[1]; Greenwalt, Cameron ^[2]; Vetter, Jeffrey ^[1]

---

1. ORNL
2. Brigham Young University

Decomposing sparse matrices into lower and upper triangular matrices (sparse LU factorization) is a key operation in many computational scientific applications. We developed SparseLU, a sparse linear algebra library that implements a new algorithm for LU factorization on general sparse matrices. The new algorithm divides the input matrix into tiles to which OpenMP tasks are created for factorization computation, where only tiles that contain nonzero elements are computed. For comparative performance analysis, we used the reference library SuperLU. Testing was performed on synthetically generated matrices which replicate the conditions of the real-world matrices. SparseLU is able to reach a mean speedup of ~29× compared to SuperLU.

• Research Organization: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-00OR22725
• OSTI ID: 2000278
• Country of Publication: United States
• Language: English