A 3D Parallel Algorithm for QR Decomposition

被引:0
|
作者
Ballard, Grey [1 ]
Demmel, James [2 ]
Grigori, Laura [3 ]
Jacquelin, Mathias [4 ]
Knight, Nicholas [5 ]
机构
[1] Wake Forest Univ, Winston Salem, NC 27101 USA
[2] Univ Calif Berkeley, Berkeley, CA 94720 USA
[3] INRIA Paris Rocquencourt, Paris, France
[4] Lawrence Berkeley Natl Lab, Berkeley, CA USA
[5] NYU, New York, NY USA
基金
美国国家科学基金会; 欧盟地平线“2020”;
关键词
communication-avoiding algorithms; bandwidth-latency tradeoffs; communication-memory tradeoffs; COLLECTIVE COMMUNICATION; MATRIX; REPRESENTATION;
D O I
10.1145/3210377.3210415
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Interprocessor communication often dominates the runtime of large matrix computations. We present a parallel algorithm for computing QR decompositions whose bandwidth cost (communication volume) can be decreased at the cost of increasing its latency cost (number of messages). By varying a parameter to navigate the bandwidth/latency tradeoff, we can tune this algorithm for machines with different communication costs.
引用
收藏
页码:55 / 65
页数:11
相关论文
共 50 条
  • [1] A parallel domain decomposition algorithm for 3D turbulence modeling
    Barwolff, G
    Schwandt, H
    ZEITSCHRIFT FUR ANGEWANDTE MATHEMATIK UND MECHANIK, 1997, 77 : S511 - S512
  • [2] ON THE QR ALGORITHM AND UPDATING THE SVD AND THE URV DECOMPOSITION IN PARALLEL
    MOONEN, M
    VANDOOREN, P
    VANPOUCKE, F
    LINEAR ALGEBRA AND ITS APPLICATIONS, 1993, 188 : 549 - 568
  • [3] The research and application of the parallel algorithm for QR decomposition of matrix
    Yang, Aimin
    Liu, Chunfeng
    Ma, Xinghua
    Ji, Nan
    An, Min
    DCABES 2006 Proceedings, Vols 1 and 2, 2006, : 29 - 31
  • [4] The parallel algorithm of QR decomposition of matrix in cluster system
    Liu, Chunfeng
    Yang, Aimin
    Chang, Jincai
    He, Yali
    Yan, Shaohong
    Advances in Matrix Theory and Applications, 2006, : 261 - 264
  • [5] A 3D parallel shrinking algorithm
    Palágyi, Kálmán
    Acta Cybernetica, 2001, 15 (02): : 201 - 211
  • [6] A 3D Skeletonization Algorithm for 3D Mesh Models Using a Partial Parallel 3D Thinning Algorithm and 3D Skeleton Correcting Algorithm
    Jin, Xun
    Kim, Jongweon
    APPLIED SCIENCES-BASEL, 2017, 7 (02):
  • [7] Fault tolerant QR-decomposition algorithm and its parallel implementation
    Maslennikow, O
    Kaniewski, J
    Wyrzykowski, R
    EURO-PAR '98 PARALLEL PROCESSING, 1998, 1470 : 798 - 803
  • [8] A parallel algorithm for 3D dislocation dynamics
    Wang, Zhiqiang
    Ghoniem, Nasr
    Swaminarayan, Sriram
    LeSar, Richard
    JOURNAL OF COMPUTATIONAL PHYSICS, 2006, 219 (02) : 608 - 621
  • [9] Fast dual-domain reduced-rank algorithm for 3D deblending via randomized QR decomposition
    Cheng, Jinkun
    Sacchi, Mauricio D.
    GEOPHYSICS, 2016, 81 (01) : V89 - V101
  • [10] Iterative QR Decomposition-Based Parallel Diversity Noncoherent Detection Algorithm
    Jieling Wang
    Bin Zhou
    Mao Zhao
    Wireless Personal Communications, 2023, 132 : 2823 - 2838