Difference between revisions of "Applications/Hypre"
m (Pysdlb moved page Hypre to Applications/Hypre without leaving a redirect) |
m |
||
| Line 19: | Line 19: | ||
===Compilation=== | ===Compilation=== | ||
| − | ''hypre'' is a library | + | ''hypre'' is a library (required at compilation time). |
A typical C++ code snippet requiring '''hypre''' (source Github) is shown below: | A typical C++ code snippet requiring '''hypre''' (source Github) is shown below: | ||
Revision as of 08:49, 24 April 2017
Application Details
- Description: a software library of high performance preconditioners and solvers for the solution of large, sparse linear systems of equations on massively parallel computers
- Version: 2.11.0 (compiled with intel)
- Module: hypre/intel/2.11.0 and hypre/intel/2.11.0b
- Licence: GNU Lesser General Public License
Usage Examples
hypre, a software library of high performance pre conditioners and solvers for the solution of large, sparse linear systems of equations on massively parallel computers. The hypre library was created with the primary goal of providing users with advanced parallel pre conditioners. The library features parallel multi-grid solvers for both structured and unstructured grid problems. For ease of use, these solvers are accessed from the application code via hypre’s conceptual linear system interfaces, which allow a variety of natural problem descriptions.
Module
[username@login01 ~]$ module add hypre/intel/2.11.0
Compilation
hypre is a library (required at compilation time).
A typical C++ code snippet requiring hypre (source Github) is shown below:
Description: This code solves a system corresponding to a discretization
of the Laplace equation -Delta u = 1 with zero boundary
conditions on the unit square. The domain is split into
a n x n grid of quadrilateral elements and each processors
owns a horizontal strip of size m x n, where m = n/nprocs. We
use bilinear finite element discretization, so there are
nodes (vertices) that are shared between neighbouring
processors. The Finite Element Interface is used to assemble
the matrix and solve the problem. Nine different solvers are
available.
*/
#include <math.h>
#include <iostream>
#include <fstream>
#include "_hypre_utilities.h"
#include "LLNL_FEI_Impl.h"
using namespace std;
#include "vis.c"
int main(int argc, char *argv[])
{
int i, j, k;
int nprocs, mypid;
int n, m, offset;
double h;
int solverID;
int vis;
// Initialize MPI
MPI_Init(&argc, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &nprocs);
MPI_Comm_rank(MPI_COMM_WORLD, &mypid);
...
The module provides the library 'libHYPRE.a' and the include files:
_hypre_IJ_mv.h _hypre_parcsr_ls.h _hypre_parcsr_mv.h _hypre_sstruct_ls.h _hypre_sstruct_mv.h _hypre_struct_ls.h _hypre_struct_mv.h _hypre_utilities.hlobpcg.h cfei.h cfei_hypre.h cfei-hypre.h Cnames.h colamd.h csr_block_matrix.h csr_matmultivec.h distributed_matrix.hhypre_cfei.h fei_bool.h fei_Data.hpp fei_defs.h fei_LinearSystemCore.hpp fei_Lookup.hpp fei_mpi.h fortran.h fortran_matrix.h HYPRE.h HYPRE_config.h HYPRE_DistributedMatrixPilutSolver_protos.h HYPRE_DistributedMatrixPilutSolver_types.h HYPRE_error_f.h HYPRE_FEI_includes.h hypre_hopscotch_hash.h HYPRE_IJ_mv.h HYPRE_krylov.h HYPRE_LinSysCore.h HYPRE_lobpcg.h HYPRE_matrix_matrix_protos.h HYPRE_MatvecFunctions.h HYPRE_parcsr_ls.h HYPRE_parcsr_mv.h HYPRE_seq_mv.h HYPRE_sstruct_ls.h HYPRE_sstruct_mv.h HYPRE_struct_ls.h HYPRE_struct_mv.h HYPRE_utilities.h HYPREf.h interpreter.h krylov.h LLNL_FEI_Fei.h LLNL_FEI_Impl.h LLNL_FEI_LSCore.h LLNL_FEI_Matrix.h LLNL_FEI_Solver.h multivector.h old_colamd.h par_csr_block_matrix.h par_csr_matmultivec.h par_csr_pmvcomm.h par_multivector.h seq_multivector.h seq_mv.h slu_cdefs.h slu_Cnames.h slu_dcomplex.h slu_ddefs.h slu_scomplex.h slu_sdefs.h slu_util.h slu_zdefs.h supermatrix.h temp_multivector.h
Further Information
|