add large vtu test

A a test that writes a .vtu file with parallel IO that is larger than
4GB.

diff --git a/tests/distributed_grids/large_vtu_01.cc b/tests/distributed_grids/large_vtu_01.cc
new file mode 100644 (file)
index 0000000..18b6131
--- /dev/null
+++ b/tests/distributed_grids/large_vtu_01.cc
@@ -0,0 +1,131 @@
+/* ---------------------------------------------------------------------
+ *
+ * Copyright (C) 2021 by the deal.II authors
+ *
+ * This file is part of the deal.II library.
+ *
+ * The deal.II library is free software; you can use it, 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 2.1 of the License, or (at your option) any later version.
+ * The full text of the license can be found in the file LICENSE.md at
+ * the top level directory of deal.II.
+ *
+ * ---------------------------------------------------------------------
+ */
+
+// Test writing of large vtu files with a total file size above 4 GB
+
+// This test is running a tiny version by default. Set the following
+// flag to true to run a test that generates a file larger than 4GB
+// when running with 2 MPI ranks. The total file size is about
+// 7 GB. Warning, you need quite a bit of RAM and patience to run this.
+const bool run_big = false;
+
+#include <deal.II/base/conditional_ostream.h>
+#include <deal.II/base/function.h>
+#include <deal.II/base/function_lib.h>
+#include <deal.II/base/index_set.h>
+#include <deal.II/base/quadrature_lib.h>
+#include <deal.II/base/utilities.h>
+
+#include <deal.II/distributed/grid_refinement.h>
+#include <deal.II/distributed/solution_transfer.h>
+#include <deal.II/distributed/tria.h>
+
+#include <deal.II/dofs/dof_accessor.h>
+#include <deal.II/dofs/dof_handler.h>
+#include <deal.II/dofs/dof_tools.h>
+
+#include <deal.II/fe/fe_dgq.h>
+#include <deal.II/fe/fe_q.h>
+#include <deal.II/fe/fe_values.h>
+
+#include <deal.II/grid/grid_generator.h>
+#include <deal.II/grid/tria_accessor.h>
+#include <deal.II/grid/tria_iterator.h>
+
+#include <deal.II/lac/la_parallel_vector.h>
+#include <deal.II/lac/sparsity_tools.h>
+#include <deal.II/lac/vector.h>
+
+#include <deal.II/numerics/data_out.h>
+#include <deal.II/numerics/error_estimator.h>
+#include <deal.II/numerics/vector_tools.h>
+
+#include <fstream>
+#include <iostream>
+
+#include "../tests.h"
+
+template <int dim>
+void
+run()
+{
+  MPI_Comm                                  mpi_communicator = MPI_COMM_WORLD;
+  parallel::distributed::Triangulation<dim> triangulation(mpi_communicator);
+
+  GridGenerator::subdivided_hyper_cube(triangulation, 1);
+  unsigned int n_cycles_global = (run_big) ? 4 : 1;
+  triangulation.refine_global(n_cycles_global);
+
+  deallog << "n_global_active_cells: " << triangulation.n_global_active_cells()
+          << " n_global_levels: " << triangulation.n_global_levels()
+          << std::endl;
+
+  const unsigned int n_vectors = 1;
+
+  FE_DGQ<dim>     fe(2);
+  DoFHandler<dim> dof_handler(triangulation);
+  dof_handler.distribute_dofs(fe);
+  // Make FE vector
+  IndexSet locally_relevant_dofs;
+  DoFTools::extract_locally_relevant_dofs(dof_handler, locally_relevant_dofs);
+
+  using VectorType = LinearAlgebra::distributed::Vector<double>;
+  VectorType global_dof_vector;
+  global_dof_vector.reinit(dof_handler.locally_owned_dofs(),
+                           locally_relevant_dofs,
+                           MPI_COMM_WORLD);
+
+  VectorTools::interpolate(dof_handler,
+                           Functions::SquareFunction<dim>(),
+                           global_dof_vector);
+
+  {
+    DataOut<dim> data_out;
+    data_out.attach_dof_handler(dof_handler);
+    for (unsigned int i = 0; i < n_vectors; ++i)
+      data_out.add_data_vector(
+        dof_handler,
+        global_dof_vector,
+        std::vector<std::string>(1, "data" + Utilities::to_string(i)));
+
+    data_out.build_patches((run_big) ? 42 : 2);
+    data_out.write_vtu_in_parallel("base.vtu", MPI_COMM_WORLD);
+  }
+
+  if (dealii::Utilities::MPI::this_mpi_process(MPI_COMM_WORLD) == 0)
+    {
+      // Let's print 10 lines from vtu (exclude the header and the data):
+      std::ifstream in("base.vtu");
+      for (int i = 0; i < 10; ++i)
+        {
+          std::string line;
+          std::getline(in, line);
+          if (line[0] == '<')
+            deallog << line << '\n';
+        }
+    }
+  deallog << "OK" << std::endl;
+}
+
+
+int
+main(int argc, char *argv[])
+{
+  Utilities::MPI::MPI_InitFinalize mpi_initialization(argc, argv, 1);
+  MPILogInitAll                    all;
+
+  run<3>();
+}

diff --git a/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output b/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output
new file mode 100644 (file)
index 0000000..40c58b1
--- /dev/null
+++ b/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output
@@ -0,0 +1,12 @@
+
+DEAL:0::n_global_active_cells: 8 n_global_levels: 2
+DEAL:0::<?xml version="1.0" ?> 
+<!-- 
+<VTKFile type="UnstructuredGrid" version="0.1" compressor="vtkZLibDataCompressor" byte_order="LittleEndian">
+<UnstructuredGrid>
+<Piece NumberOfPoints="216" NumberOfCells="64" >
+OK
+
+DEAL:1::n_global_active_cells: 8 n_global_levels: 2
+DEAL:1::OK
+

diff --git a/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output.big b/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output.big
new file mode 100644 (file)
index 0000000..e0ffffe
--- /dev/null
+++ b/tests/distributed_grids/large_vtu_01.with_p4est=true.mpirun=4.output.big
@@ -0,0 +1,20 @@
+
+DEAL:0::n_global_active_cells: 4096 n_global_levels: 5
+DEAL:0::<?xml version="1.0" ?> 
+<!-- 
+<VTKFile type="UnstructuredGrid" version="0.1" compressor="vtkZLibDataCompressor" byte_order="LittleEndian">
+<UnstructuredGrid>
+<Piece NumberOfPoints="81415168" NumberOfCells="75866112" >
+OK
+
+DEAL:1::n_global_active_cells: 4096 n_global_levels: 5
+DEAL:1::OK
+
+DEAL:2::n_global_active_cells: 4096 n_global_levels: 5
+DEAL:2::OK
+
+DEAL:3::n_global_active_cells: 4096 n_global_levels: 5
+DEAL:3::OK
+
+
+