Tpetra, and the use of generic programming in scientific computing

被引：0

作者：

Baker, C. G. ^{[1
]}

Heroux, M. A. ^{[2
]}

机构：

[1] Oak Ridge Natl Lab, Computat Engn & Energy Sci Grp, Oak Ridge, TN 37831 USA

[2] Sandia Natl Labs, Dept Scalable Algorithms, Albuquerque, NM 87185 USA

来源：

SCIENTIFIC PROGRAMMING | 2012年 / 20卷 / 02期

关键词：

Generic programming; scientific computing; template metaprogramming; shared-memory programming; distributed-memory programming; many-core computing;

D O I：

10.1155/2012/693861

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

We present Tpetra, a Trilinos package for parallel linear algebra primitives implementing the Petra object model. We describe Tpetra's design, based on generic programming via C++ templated types and template metaprogramming. We discuss some benefits of this approach in the context of scientific computing, with illustrations consisting of code and notable empirical results.

引用

页码：115 / 128

页数：14

共 26 条

[1]

[Anonymous], 2010, EPETRA 10 6 RELEASE

[2]

[Anonymous], 2010, Thrust: A parallel template library

[3]

Bailey D.H., 2002, 53651 LBNL

[4]

Bailey D.H., 2011, HIGH PRECISION SOFTW

[5] Anasazi Software for the Numerical Solution of Large-Scale Eigenvalue Problems [J].

Baker, C. G. ;

Hetmaniuk, U. L. ;

Lehoucq, R. B. ;

Thornquist, H. K. .

ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2009, 36 (03)

[6]

Baker C.G., 2009, SIAM ANN M JUL

[7]

Baker C.G., 2011, TPETRA RECURSIVE MIX

[8]

Baker C.G., 2011, TPETRA CONJUGATE GRA

[9]

Baker C.G., 2010, SIAM C PAR PROC SCI

[10]

Baker C.G., 2010, PARALLEL PROGRAMMING

← 1 2 3 →