Performance optimization and modeling of blocked sparse kernels

被引：20

作者：

Buttari, Alfredo ^{[1
]}

Eijkhout, Victor

Langou, Julien

Filippone, Salvatore

机构：

[1] Univ Tennessee, Innovat Comp Lab, Knoxville, TN 37996 USA

[2] Univ Texas, Texas Adv Comp Lab, Austin, TX 78712 USA

[3] Univ Colorado, Dept Math Sci, Denver, CO 80202 USA

[4] Univ Colorado, Hlth Sci Ctr, Denver, CO USA

[5] Univ Roma Tor Vergata, Rome, Italy

来源：

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS | 2007年 / 21卷 / 04期

关键词：

optimization; sparse; matrix-vector product; blocking; self-adaptivity;

D O I：

10.1177/1094342007083801

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

We present a method for automatically selecting optimal implementations of sparse matrix-vector operations. Our software "AcCELS" (Accelerated Compress-storage Elements for Linear Solvers) involves a setup phase that probes machine characteristics, and a run-time phase where stored characteristics are combined with a measure of the actual sparse matrix to find the optimal kernel implementation. We present a performance model that is shown to be accurate over a large range of matrices.

引用

页码：467 / 484

页数：18

共 13 条

[1] A fully asynchronous multifrontal solver using distributed dynamic scheduling
Amestoy, PR
Duff, IS
L'Excellent, JY
Koster, J
[J]. SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2001, 23 (01) : 15 - 41
[2] Barrett R., 1994, Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods, V2nd ed.
[3] Davis MJ, 2001, ADV PHYSIOL EDUC, V25, P1
[4] Self-adapting numerical software for next generation applications
Dongarra, J
Eijkhout, V
[J]. INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2003, 17 (02) : 125 - 131
[5] PSBLAS: A library for parallel linear algebra computation on sparse matrices
Filippone, S
Colajanni, M
[J]. ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2000, 26 (04): : 527 - 550
[6] Sparsity: Optimization framework for sparse matrix kernels
Im, EJ
Yelick, K
Vuduc, R
[J]. INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2004, 18 (01) : 135 - 158
[7] IM EJ, 1998, WORKSH PROF FEEDB DI
[8] LI XS, 1996, THESIS U CALIFORNIA
[9] PINAR A, 1999, P SUPERCOMPUTING 99
[10] Toledo S, 1997, P 8 SIAM C PAR PROC

← 1 2 →