Geodesic
TIM-parallelization in a shared memory model using OpenMP interface
Numerical methods and programming, Tome 8 (2007) no. 1, pp. 130-137

Voir la notice de l'article provenant de la source Math-Net.Ru

The computational TIM technique is devoted to solving time-dependent multidimensional continuum mechanics problems using arbitrarily unstructured Lagrangian grids. The TIM technique allows one to solve 2D problems (TIM-2D) in cylindrical and Cartesian coordinates and 3D problems (TIM-3D) in Cartesian coordinates. To reduce run-times, TIM implements parallelization in a shared memory model using OpenMP interface. Program parallelization is performed by OpenMP parallelization directives added to each loop whose iterations may be independent of each other. Each loop is parallelized independently. Computational modules for gas dynamics, elasticity-plasticity, magnetic hydrodynamics, two-flow and two-temperature behavior, and maintenance of grids, as well as a number of support procedures are parallelized. In some cases we had to revise algorithms used for sequential computations. Parallelization is implemented for computational modules requiring 99 The paper was prepared on the basis of the authors' report at the International Conference on Parallel Computing Technologies (PaVT-2007; http://agora.guru.ru/pavt).
Mots-clés : методика ТИМ; распараллеливание на общей памяти; интерфейс OpenMP; нерегулярные сетки; параллельные алгоритмы TIM technique
Keywords: shared memory parallelization, OpenMP interface, irregular grids, parallel algorithms. 
@article{VMP_2007_8_1_a15,
     author = {A. A. Voropinov and I. G. Novikov and I. V. Sobolev and S. S. Sokolov},
     title = {TIM-parallelization in a shared memory model using {OpenMP} interface},
     journal = {Numerical methods and programming},
     pages = {130--137},
     publisher = {mathdoc},
     volume = {8},
     number = {1},
     year = {2007},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2007_8_1_a15/}
}
TY  - JOUR
AU  - A. A. Voropinov
AU  - I. G. Novikov
AU  - I. V. Sobolev
AU  - S. S. Sokolov
TI  - TIM-parallelization in a shared memory model using OpenMP interface
JO  - Numerical methods and programming
PY  - 2007
SP  - 130
EP  - 137
VL  - 8
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2007_8_1_a15/
LA  - ru
ID  - VMP_2007_8_1_a15
ER  - 
%0 Journal Article
%A A. A. Voropinov
%A I. G. Novikov
%A I. V. Sobolev
%A S. S. Sokolov
%T TIM-parallelization in a shared memory model using OpenMP interface
%J Numerical methods and programming
%D 2007
%P 130-137
%V 8
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2007_8_1_a15/
%G ru
%F VMP_2007_8_1_a15
A. A. Voropinov; I. G. Novikov; I. V. Sobolev; S. S. Sokolov. TIM-parallelization in a shared memory model using OpenMP interface. Numerical methods and programming, Tome 8 (2007) no. 1, pp. 130-137. http://geodesic.mathdoc.fr/item/VMP_2007_8_1_a15/