Some Features of the CUDA Implementation of the Semi-Lagrangian Method for the Advection Problem

被引：3

作者：

Efremov, A. ^{[1
]}

Karepova, E. ^{[1
,2
]}

Vyatkin, A. ^{[1
,2
]}

机构：

[1] SB RAS, Inst Computat Modeling, Akademgorodok, Krasnoyarsk 660036, Russia

[2] Siberian Fed Univ, Krasnoyarsk 660041, Russia

来源：

APPLICATION OF MATHEMATICS IN TECHNICAL AND NATURAL SCIENCES (AMITANS'15) | 2015年 / 1684卷

关键词：

D O I：

10.1063/1.4934328

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

In the paper the semi-Lagrangian method is considered in the context of its implementation with the CUDA technology. We have scrutinized the bottleneck of our sequential algorithm; its parallel versions are studied in detail; and the main reason of poor CUDA performance is clarified. As the result, we revise the computation of partial integrals in order to improve the efficiency of the algorithm. Numerical experiments demonstrate good CUDA performance of the revised version of the algorithm.

引用

页数：10

共 15 条

[1]

Anderson John David, 1995, Computational Fluid Dynamics, V206

[2]

[Anonymous], 2002, Cambridge Texts in Applied Mathematics, DOI [10.1017/CBO9780511791253, DOI 10.1017/CBO9780511791253]

[3]

[Anonymous], 2010, TESLA C2050 TESLA C2

[4]

[Anonymous], 1999, Texts in Applied Mathematics

[5]

Behrens J., 1997, NOTES NUMERICAL FLUI, V59, P49

[6]

Carlini E., 2006, NUMERICAL MATH ADV A, P732

[7] Error Estimates for Triangular and Tetrahedral Finite Elements in Combination with a Trajectory Approximation of the First Derivatives for Advection-Diffusion Equations [J].

Chen, H. ;

Lin, Q. ;

Shaidurov, V. V. ;

Zhou, J. .

NUMERICAL ANALYSIS AND APPLICATIONS, 2011, 4 (04) :345-362

[8]

Efremov A., 2014, J APPL MATH, V2014

[9] A Semi-Lagrangian Method for a Fokker-Planck Equation Describing Fiber Dynamics [J].

Klar, Axel ;

Reuterswaerd, Philip ;

Seaid, Mohammed .

JOURNAL OF SCIENTIFIC COMPUTING, 2009, 38 (03) :349-367

[10]

Morton K.W., 1996, Numerical Solution of Convection-Diffusion Problems

← 1 2 →