A parallel structured banded DC algorithm for symmetric eigenvalue problems

被引：2

作者：

Li, Shengguo ^{[1
]}

Liao, Xia ^{[1
]}

Lu, Yutong ^{[2
,3
]}

Roman, Jose E. ^{[4
]}

Yue, Xiaoqiang ^{[5
]}

机构：

[1] Natl Univ Def Technol, Coll Comp Sci, Changsha 410073, Peoples R China

[2] Sun Yatsen Univ, Natl Supercomp Ctr Guangzhou, Guangzhou 510006, Peoples R China

[3] Sun Yatsen Univ, Sch Data & Comp Sci, Guangzhou 510006, Peoples R China

[4] Univ Politecn Valencia, D Sistemes Informat & Comp, Cami Vera S-N, Valencia 46022, Spain

[5] Xiangtan Univ, Hunan Key Lab Computat & Simulat Sci & Engn, Key Lab Intelligent Comp & Informat Proc, Minist Educ, Xiangtan 411105, Peoples R China

来源：

CCF TRANSACTIONS ON HIGH PERFORMANCE COMPUTING | 2023年 / 5卷 / 02期

关键词：

ScaLAPACK; Divide-and-conquer; PSMMA; PBSDC; Distributed-memory parallel algorithm; RANK-ONE MODIFICATION; CONQUER ALGORITHM; DIVIDE; FRAMEWORK;

D O I：

10.1007/s42514-022-00117-9

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, a novel parallel structured divide-and-conquer (DC) algorithm is proposed for symmetric banded eigenvalue problems, denoted by PBSDC, which modifies the classical parallel banded DC (PBDC) algorithm by reducing its computational cost. The main tool that PBSDC uses is a parallel structured matrix multiplication algorithm (PSMMA), which can be much faster than the general dense matrix multiplication ScaLAPACK routine PDGEMM. Numerous experiments have been performed on Tianhe-2 supercomputer to compare PBSDC with PBDC and ELPA. For matrices with few deflations, PBSDC can be much faster than PBDC since computations are saved. For matrices with many deflations and/or small bandwidths, PBSDC can be faster than the tridiagonalization-based DC implemented in LAPACK and ELPA. However, PBSDC would become slower than ELPA for matrices with relatively large bandwidths.

引用

页码：116 / 128

页数：13

共 16 条

[1] A parallel structured banded DC algorithm for symmetric eigenvalue problems
Shengguo Li
Xia Liao
Yutong Lu
Jose E. Roman
Xiaoqiang Yue
CCF Transactions on High Performance Computing, 2023, 5 : 116 - 128
[2] A Parallel Structured Divide-and-Conquer Algorithm for Symmetric Tridiagonal Eigenvalue Problems
Liao, Xia
Li, Shengguo
Lu, Yutong
Roman, Jose E.
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2021, 32 (02) : 367 - 378
[3] A parallel divide and conquer algorithm for the symmetric eigenvalue problem on distributed memory architectures
Tisseur, F
Dongarra, J
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1999, 20 (06) : 2223 - 2236
[4] Parallel solution of partial symmetric eigenvalue problems from electronic structure calculations
Auckenthaler, T.
Blum, V.
Bungartz, H. -J.
Huckle, T.
Johanni, R.
Kraemer, L.
Lang, B.
Lederer, H.
Willems, P. R.
PARALLEL COMPUTING, 2011, 37 (12) : 783 - 794
[5] A PARALLEL ALGORITHM FOR THE NONSYMMETRIC EIGENVALUE PROBLEM
DONGARRA, JJ
SIDANI, M
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1993, 14 (03) : 542 - 569
[6] Developing algorithms and software for the parallel solution of the symmetric eigenvalue problem
Auckenthaler, T.
Bungartz, H. -J.
Huckle, T.
Kraemer, L.
Lang, B.
Willems, P.
JOURNAL OF COMPUTATIONAL SCIENCE, 2011, 2 (03) : 272 - 278
[7] DEFLATION FOR THE SYMMETRIC ARROWHEAD AND DIAGONAL-PLUS-RANK-ONE EIGENVALUE PROBLEMS
Barlow, Jesse L.
Eisenstat, Stanley C.
Stor, Nevena Jakovcevic
Slapnicar, Ivan
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2022, 43 (02) : 681 - 709
[8] A parallel hybrid banded system solver: the SPIKE algorithm
Polizzi, E
Sameh, AH
PARALLEL COMPUTING, 2006, 32 (02) : 177 - 194
[9] Tridiagonalization of a dense symmetric matrix on multiple GPUs and its application to symmetric eigenvalue problems
Yamazaki, Ichitaro
Dong, Tingxing
Solca, Raffaele
Tomov, Stanimire
Dongarra, Jack
Schulthess, Thomas
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2014, 26 (16) : 2652 - 2666
[10] Algorithm 826: A parallel eigenvalue routine for complex Hessenberg matrices
Fahey, MR
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2003, 29 (03): : 326 - 336

← 1 2 →