Recent Commits
Commit | Author | Details | Committed | ||||
---|---|---|---|---|---|---|---|
ab9904f84556 | pastewka | TST: Updated Python tests to also test FFTWMPI and PFFT (in serial). | Mar 29 2018 | ||||
78e53a9268db | pastewka | ENH: Python bindings and thin dispatch wrapper for FFT engines. | Mar 29 2018 | ||||
afdd9fe06f44 | pastewka | BUG: We can't really call fftw_cleanup until we know that FFTW will no longer… | Mar 29 2018 | ||||
89e042fece5c | D174 | pastewka | ENH: MPI parallelization | Mar 27 2018 | |||
3d9fd45f1098 | D171 | junge | Complete draft of coding convention | Mar 26 2018 | |||
dad5c7939db6 | D168 | RLeute | Implementation of a ProjectionError for even grids, need discussion | Mar 22 2018 | |||
45c48a4eea3e | junge | Merge branch 'arcpatch-D169' into feat/documentation | Mar 17 2018 | ||||
3795c8590d33 | pastewka | DOC: Added documentation for undocumented members. | Mar 16 2018 | ||||
9137a5a94322 | junge | Made the communicator effc++ compatible | Mar 16 2018 | ||||
fad17c728490 | D169 | junge | ENH: First implementation of MPI parallelization. | Mar 16 2018 | |||
acb656fa2706 | D169 | junge/pastewka | ENH: First implementation of MPI parallelization. | Mar 16 2018 | |||
0f03cbf18341 | pastewka | WIP: Updated language bindings to reflect recent changes in API. | Mar 16 2018 | ||||
b3d3315e59c7 | pastewka | WIP: HAVE_MPI should have been WITH_MPI | Mar 16 2018 | ||||
679cc5d1e141 | pastewka | WIP: Enabled and updated MPI parallel patch test. | Mar 16 2018 | ||||
2c848fe691c0 | pastewka | WIP: Added convenience functions to create parallel cells. | Mar 16 2018 |
README
µSpectre
Copyright © 2018 Till Junge
µSpectre is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3, or (at your option) any later version.
µSpectre 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 General Public License for more details.
You should have received a copy of the GNU General Public License along with GNU Emacs; see the file COPYING. If not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
This README contains only a small quick start guide. Please refer to the full documentation for more help.
Building µSpectre
µSpectre is a CMake project that uses C++14. It depends on the Boost unit test framework for testing and uses uses python3 as secondary API. You will need a modern C++ compiler (µSpectre was tested with gcc-6 , gcc-7, clang-4 and clang5) and CMake version 3.7.0 or higher.
Compilation
$ git clone https://c4science.ch/source/muSpectre.git $ cd build $ cmake -DCMAKE_BUILD_TYPE=Release .. $ make
µSpectre makes use of expression templates, as a result, production code must be compiled with the CMAKE_BUILD_TYPE flag set to Release in order to get performance (under gcc, non-optimised code is about 50 times slower than the release). Be careful with parallel compilation: compiling µSpectre is quite memory-hungry because of the use of expression templates, a parallel make -j requires currently about 10 GB of RAM under GCC.
Simple usage example
The following is a simple example for using µSpectre through its convenient Python interface
#!/usr/bin/env python3 import numpy as np import muSpectre as µ # setting the geometry resolution = [51, 51] center = np.array([r//2 for r in resolution]) incl = resolution[0]//5 lengths = [7., 5.] formulation = µ.Formulation.small_strain # creating the periodic cell rve = µ.SystemFactory(resolution, lengths, formulation) hard = µ.material.MaterialLinearElastic1_2d.make( rve, "hard", 10e9, .33) soft = µ.material.MaterialLinearElastic1_2d.make( rve, "soft", 70e9, .33) # assign a material to each pixel for i, pixel in enumerate(rve): if np.linalg.norm(center - np.array(pixel),2)<incl: hard.add_pixel(pixel) else: soft.add_pixel(pixel) tol = 1e-5 cg_tol = 1e-8 # set macroscopic strain Del0 = np.array([[.0, .0], [0, .03]]) if formulation == µ.Formulation.small_strain: Del0 = .5*(Del0 + Del0.T) maxiter = 401 verbose = 2 solver = µ.solvers.SolverCG(rve, cg_tol, maxiter, verbose=False) r = µ.solvers.newton_cg(rve, Del0, solver, tol, verbose) print("nb of {} iterations: {}".format(solver.name(), r.nb_fev))
You can find more examples using both the python and the c++ interface in the bin/ and tests folder.