Fast Sparse Cholesky Decomposition and Inversion using Nested Dissection Matrix Reordering

被引:15
|
作者
Brandhorst, Kai [1 ]
Head-Gordon, Martin [1 ,2 ]
机构
[1] Univ Calif Berkeley, Dept Chem, Berkeley, CA 94720 USA
[2] Lawrence Berkeley Natl Lab, Div Chem Sci, Berkeley, CA USA
关键词
ELECTRONIC-STRUCTURE CALCULATIONS; PLESSET CORRELATION-ENERGY; DEGREE ORDERING ALGORITHM; DENSITY-MATRIX; NULL SPACE; DIRECT OPTIMIZATION; TENSOR FORMULATION; ELIMINATION TREES; MOLECULAR-SYSTEMS; HARTREE-FOCK;
D O I
10.1021/ct100618s
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Here we present an efficient, yet nonlinear scaling, algorithm for the computation of Cholesky factors of sparse symmetric positive definite matrices and their inverses. The key feature of this implementation is the separation of the task into an algebraic and a numeric part. The algebraic part of the algorithm attempts to find a reordering of the rows and columns which preserves at least some degree of sparsity and afterward determines the exact nonzero structure of both the Cholesky factor and its corresponding inverse. It is based on graph theory and does not involve any kind of numerical thresholding. This preprocessing then allows for a very efficient implementation of the numerical factorization step. Furthermore this approach even allows use of highly optimized dense linear algebra kernels which leads to yet another performance boost. We will show some illustrative timings of our sparse code and compare it to the standard library implementation and a recent sparse implementation using thresholding. We conclude with some comments on how to deal with positive semidefinite matrices.
引用
收藏
页码:351 / 368
页数:18
相关论文
共 50 条
  • [41] Fast iterative solution of integral equation with matrix decomposition algorithm and multilevel simple sparse method
    Hu, X. Q.
    Xu, Y.
    Chen, R. S.
    IET MICROWAVES ANTENNAS & PROPAGATION, 2011, 5 (13) : 1583 - 1588
  • [42] A fast global nodewise mass matrix inversion framework tailored for sparse block-diagonal systems
    Rekatsinas, C. S.
    THIN-WALLED STRUCTURES, 2022, 172
  • [43] Fast reconstruction algorithms for optical tomography using sparse matrix representations
    Cao, Guangzhi
    Bouman, Charles A.
    Webb, Kevin J.
    2007 4TH IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING : MACRO TO NANO, VOLS 1-3, 2007, : 912 - 915
  • [44] Fast and accurate T2 mapping using Bloch simulations and low-rank plus sparse matrix decomposition
    Daniel, Grzeda
    Meirav, Galun
    Noam, Omer
    Tamar, Blumenfeld-Katzir
    Dvir, Radunsky
    Ricardo, Otazo
    Noam, Ben-Eliezer
    MAGNETIC RESONANCE IMAGING, 2023, 98 : 66 - 75
  • [45] Fast randomized matrix and tensor interpolative decomposition using CountSketch
    Malik, Osman Asif
    Becker, Stephen
    ADVANCES IN COMPUTATIONAL MATHEMATICS, 2020, 46 (06)
  • [46] Fast randomized matrix and tensor interpolative decomposition using CountSketch
    Osman Asif Malik
    Stephen Becker
    Advances in Computational Mathematics, 2020, 46
  • [47] A fast parallel Gauss Jordan algorithm for matrix inversion using CUDA
    Sharma, Girish
    Agarwala, Abhishek
    Bhattacharya, Baidurya
    COMPUTERS & STRUCTURES, 2013, 128 : 31 - 37
  • [48] Broadband Analysis of Microwave Devices Using Asymptotic Waveform Evaluation and Nested Matrix Decomposition
    Wan, Ting
    Li, Mengzhe
    Tang, Benliu
    2017 IEEE SIXTH ASIA-PACIFIC CONFERENCE ON ANTENNAS AND PROPAGATION (APCAP), 2017,
  • [49] SCOPE: signal compensation for low-rank plus sparse matrix decomposition for fast parameter mapping
    Zhu, Yanjie
    Liu, Yuanyuan
    Ying, Leslie
    Peng, Xi
    Wang, Yi-Xiang J.
    Yuan, Jing
    Liu, Xin
    Liang, Dong
    PHYSICS IN MEDICINE AND BIOLOGY, 2018, 63 (18):
  • [50] Fast iterative image reconstruction using sparse matrix factorization with GPU acceleration
    Zhou, Jian
    Qi, Jinyi
    MEDICAL IMAGING 2011: PHYSICS OF MEDICAL IMAGING, 2011, 7961