Homec4science
Diffusion akantu (features/30-implementing-a-checkpoint-restart-mechanism) Browse

rAKA/src/solver/solver_mumps.cc
514a38ab99effeatures/30-implementin…

Last Change
Raw File

solver_mumps.cc
View Options

/**
* @file solver_mumps.cc
*
* @author Nicolas Richart <nicolas.richart@epfl.ch>
*
* @date creation: Mon Dec 13 2010
* @date last modification: Tue Jan 19 2016
*
* @brief implem of SolverMumps class
*
* @section LICENSE
*
* Copyright (©) 2010-2012, 2014, 2015 EPFL (Ecole Polytechnique Fédérale de
* Lausanne) Laboratory (LSMS - Laboratoire de Simulation en Mécanique des
* Solides)
*
* Akantu is free software: you can redistribute it and/or modify it under the
* terms of the GNU Lesser General Public License as published by the Free
* Software Foundation, either version 3 of the License, or (at your option) any
* later version.
*
* Akantu is distributed in the hope that it will be useful, but WITHOUT ANY
* WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
* A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
* details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with Akantu. If not, see <http://www.gnu.org/licenses/>.
*
* @section DESCRIPTION
*
* @subsection Ctrl_param Control parameters
*
* ICNTL(1),
* ICNTL(2),
* ICNTL(3) : output streams for error, diagnostics, and global messages
*
* ICNTL(4) : verbose level : 0 no message - 4 all messages
*
* ICNTL(5) : type of matrix, 0 assembled, 1 elementary
*
* ICNTL(6) : control the permutation and scaling(default 7) see mumps doc for
* more information
*
* ICNTL(7) : determine the pivot order (default 7) see mumps doc for more
* information
*
* ICNTL(8) : describe the scaling method used
*
* ICNTL(9) : 1 solve A x = b, 0 solve At x = b
*
* ICNTL(10) : number of iterative refinement when NRHS = 1
*
* ICNTL(11) : > 0 return statistics
*
* ICNTL(12) : only used for SYM = 2, ordering strategy
*
* ICNTL(13) :
*
* ICNTL(14) : percentage of increase of the estimated working space
*
* ICNTL(15-17) : not used
*
* ICNTL(18) : only used if ICNTL(5) = 0, 0 matrix centralized, 1 structure on
* host and mumps give the mapping, 2 structure on host and distributed matrix
* for facto, 3 distributed matrix
*
* ICNTL(19) : > 0, Shur complement returned
*
* ICNTL(20) : 0 rhs dense, 1 rhs sparse
*
* ICNTL(21) : 0 solution in rhs, 1 solution distributed in ISOL_loc and SOL_loc
* allocated by user
*
* ICNTL(22) : 0 in-core, 1 out-of-core
*
* ICNTL(23) : maximum memory allocatable by mumps pre proc
*
* ICNTL(24) : controls the detection of "null pivot rows"
*
* ICNTL(25) :
*
* ICNTL(26) :
*
* ICNTL(27) :
*
* ICNTL(28) : 0 automatic choice, 1 sequential analysis, 2 parallel analysis
*
* ICNTL(29) : 0 automatic choice, 1 PT-Scotch, 2 ParMetis
*/
/* -------------------------------------------------------------------------- */
#include "aka_common.hh"
#if defined(AKANTU_USE_MPI)
#include "static_communicator_mpi.hh"
#include "mpi_type_wrapper.hh"
#endif
#include "solver_mumps.hh"
#include "dof_synchronizer.hh"
/* -------------------------------------------------------------------------- */
// static std::ostream & operator <<(std::ostream & stream, const DMUMPS_STRUC_C
// & _this) {
// stream << "DMUMPS Data [" << std::endl;
// stream << " + job : " << _this.job << std::endl;
// stream << " + par : " << _this.par << std::endl;
// stream << " + sym : " << _this.sym << std::endl;
// stream << " + comm_fortran : " << _this.comm_fortran << std::endl;
// stream << " + nz : " << _this.nz << std::endl;
// stream << " + irn : " << _this.irn << std::endl;
// stream << " + jcn : " << _this.jcn << std::endl;
// stream << " + nz_loc : " << _this.nz_loc << std::endl;
// stream << " + irn_loc : " << _this.irn_loc << std::endl;
// stream << " + jcn_loc : " << _this.jcn_loc << std::endl;
// stream << "]";
// return stream;
// }
__BEGIN_AKANTU__
/* -------------------------------------------------------------------------- */
SolverMumps::SolverMumps(SparseMatrix & matrix,
const ID & id,
const MemoryID & memory_id,
StaticCommunicator & comm) :
Solver(matrix, id, memory_id, comm), is_mumps_data_initialized(false), rhs_is_local(true) {
AKANTU_DEBUG_IN();
#ifdef AKANTU_USE_MPI
parallel_method = SolverMumpsOptions::_fully_distributed;
#else // AKANTU_USE_MPI
parallel_method = SolverMumpsOptions::_not_parallel;
#endif // AKANTU_USE_MPI
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
SolverMumps::~SolverMumps() {
AKANTU_DEBUG_IN();
this->destroyInternalData();
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::destroyInternalData() {
AKANTU_DEBUG_IN();
if (this->is_mumps_data_initialized) {
this->mumps_data.job = _smj_destroy; // destroy
dmumps_c(&this->mumps_data);
this->is_mumps_data_initialized = false;
}
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::initMumpsData() {
// Default Scaling
icntl(8) = 77;
// Assembled matrix
icntl(5) = 0;
/// Default centralized dense second member
icntl(20) = 0;
icntl(21) = 0;
icntl(28) = 0; // automatic choice for analysis analysis
switch (this->parallel_method) {
case SolverMumpsOptions::_serial_split:
case SolverMumpsOptions::_fully_distributed:
icntl(18) = 3; // fully distributed
this->mumps_data.nz_loc = matrix->getNbNonZero();
this->mumps_data.irn_loc = matrix->getIRN().storage();
this->mumps_data.jcn_loc = matrix->getJCN().storage();
break;
case SolverMumpsOptions::_not_parallel:
case SolverMumpsOptions::_master_slave_distributed:
icntl(18) = 0; // centralized
if (prank == 0) {
this->mumps_data.nz = matrix->getNbNonZero();
this->mumps_data.irn = matrix->getIRN().storage();
this->mumps_data.jcn = matrix->getJCN().storage();
} else {
this->mumps_data.nz = 0;
this->mumps_data.irn = NULL;
this->mumps_data.jcn = NULL;
}
break;
default:
AKANTU_DEBUG_ERROR("This case should not happen!!");
}
}
/* -------------------------------------------------------------------------- */
void SolverMumps::initialize(SolverOptions & options) {
AKANTU_DEBUG_IN();
if (SolverMumpsOptions * opt = dynamic_cast<SolverMumpsOptions *>(&options)) {
this->parallel_method = opt->parallel_method;
}
this->mumps_data.par = 1; // The host is part of computations
switch (this->parallel_method) {
case SolverMumpsOptions::_not_parallel:
break;
case SolverMumpsOptions::_master_slave_distributed:
this->mumps_data.par = 0; // The host is not part of the computations
case SolverMumpsOptions::_fully_distributed:
#ifdef AKANTU_USE_MPI
{
const StaticCommunicatorMPI & mpi_st_comm =
dynamic_cast<const StaticCommunicatorMPI &>(
communicator.getRealStaticCommunicator());
this->mumps_data.comm_fortran =
MPI_Comm_c2f(mpi_st_comm.getMPITypeWrapper().getMPICommunicator());
#endif
break;
}
case SolverMumpsOptions::_serial_split:
#ifdef AKANTU_USE_MPI
const StaticCommunicatorMPI & mpi_st_comm =
dynamic_cast<const StaticCommunicatorMPI &>(
communicator.getRealStaticCommunicator());
MPI_Comm mpi_comm = mpi_st_comm.getMPITypeWrapper().getMPICommunicator();
MPI_Comm new_mpi_comm;
Int result = MPI_Comm_split(mpi_comm, communicator.whoAmI(), 0, &new_mpi_comm);
AKANTU_DEBUG_ASSERT(result == MPI_SUCCESS, "Error in MPI_split");
this->mumps_data.comm_fortran = MPI_Comm_c2f(new_mpi_comm);
#endif
break;
}
this->mumps_data.sym = 2 * (matrix->getSparseMatrixType() == _symmetric);
if (this->parallel_method == SolverMumpsOptions::_serial_split)
this->prank = 0;
else
this->prank = communicator.whoAmI();
this->mumps_data.job = _smj_initialize; // initialize
dmumps_c(&this->mumps_data);
this->is_mumps_data_initialized = true;
/* ------------------------------------------------------------------------ */
UInt size = matrix->getSize();
if (prank == 0) {
std::stringstream sstr_rhs;
sstr_rhs << id << ":rhs";
this->rhs = &(alloc<Real>(sstr_rhs.str(), size, 1, 0.));
} else {
this->rhs = NULL;
}
this->mumps_data.nz_alloc = 0;
this->mumps_data.n = size;
/* ------------------------------------------------------------------------ */
// Output setup
if (AKANTU_DEBUG_TEST(dblTrace)) {
icntl(1) = 6;
icntl(2) = 2;
icntl(3) = 2;
icntl(4) = 4;
} else {
/// No outputs
icntl(1) = 6; // error output
icntl(2) = 0; // dignostics output
icntl(3) = 0; // informations
icntl(4) = 0; // no outputs
}
if (AKANTU_DEBUG_TEST(dblDump)) {
strcpy(this->mumps_data.write_problem, "mumps_matrix.mtx");
}
this->analysis();
// icntl(14) = 80;
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::setRHS(Array<Real> & rhs) {
if (prank == 0) {
// std::copy(rhs.storage(), rhs.storage() + this->rhs->getSize(),
// this->rhs->storage());
DebugLevel dbl = debug::getDebugLevel();
debug::setDebugLevel(dblError);
matrix->getDOFSynchronizer().gather(rhs, 0, this->rhs);
debug::setDebugLevel(dbl);
} else {
this->matrix->getDOFSynchronizer().gather(rhs, 0);
}
}
/* -------------------------------------------------------------------------- */
void SolverMumps::analysis() {
AKANTU_DEBUG_IN();
initMumpsData();
this->mumps_data.job = _smj_analyze; // analyze
dmumps_c(&this->mumps_data);
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::factorize() {
AKANTU_DEBUG_IN();
if (parallel_method == SolverMumpsOptions::_fully_distributed ||
parallel_method == SolverMumpsOptions::_serial_split)
this->mumps_data.a_loc = this->matrix->getA().storage();
else {
if (prank == 0)
this->mumps_data.a = this->matrix->getA().storage();
}
if (prank == 0) {
this->mumps_data.rhs = this->rhs->storage();
}
this->mumps_data.job = _smj_factorize; // factorize
dmumps_c(&this->mumps_data);
this->printError();
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::solve() {
AKANTU_DEBUG_IN();
if (prank == 0) {
this->mumps_data.rhs = this->rhs->storage();
}
this->mumps_data.job = _smj_solve; // solve
dmumps_c(&this->mumps_data);
this->printError();
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::solve(Array<Real> & solution) {
AKANTU_DEBUG_IN();
this->solve();
if (this->parallel_method == SolverMumpsOptions::_serial_split)
solution.copy(*(this->rhs), true);
else {
if (prank == 0) {
matrix->getDOFSynchronizer().scatter(solution, 0, this->rhs);
// std::copy(this->rhs->storage(), this->rhs->storage() +
// this->rhs->getSize(), solution.storage());
} else {
this->matrix->getDOFSynchronizer().scatter(solution, 0);
}
}
AKANTU_DEBUG_OUT();
}
/* -------------------------------------------------------------------------- */
void SolverMumps::printError() {
Int _info_v[2];
_info_v[0] = info(1); // to get errors
_info_v[1] = -info(1); // to get warnings
if (this->parallel_method != SolverMumpsOptions::_serial_split)
communicator.allReduce(_info_v, 2, _so_min);
_info_v[1] = -_info_v[1];
if (_info_v[0] < 0) { // < 0 is an error
switch (_info_v[0]) {
case -10:
AKANTU_DEBUG_ERROR("The matrix is singular");
break;
case -9: {
icntl(14) += 10;
if (icntl(14) != 90) {
// std::cout << "Dynamic memory increase of 10%" << std::endl;
AKANTU_DEBUG_WARNING(
"MUMPS dynamic memory is insufficient it will be increased of 10%");
this->analysis();
this->factorize();
this->solve();
} else {
AKANTU_DEBUG_ERROR("The MUMPS workarray is too small INFO(2)="
<< info(2) << "No further increase possible");
break;
}
}
default:
AKANTU_DEBUG_ERROR("Error in mumps during solve process, check mumps "
"user guide INFO(1) = "
<< _info_v[1]);
}
} else if (_info_v[1] > 0) {
AKANTU_DEBUG_WARNING("Warning in mumps during solve process, check mumps "
"user guide INFO(1) = "
<< _info_v[1]);
}
}
__END_AKANTU__
c4science · Help