docu: add walk-through for JUQUEEN

Author: torbjoernk

doc/source/installing.md

@@ -135,6 +135,11 @@ latest release from [GitHub][github_releases].
 
        make install
 
+### Supercomputers
+
+There are a few tweaks and special care require to get up and running on supercomputers.
+We have compiled a few working walk-throughs at \subpage page_supercomputers .
+
 ## Building with vanilla make
 
 A sample `Makefile` is included in the `advection_diffusion` example.  This may be of particular

doc/source/supercomputers.md

@@ -0,0 +1,69 @@
+# Usage on Supercomputers                                                     {#page_supercomputers}
+
+## JUQUEEN
+
+There are a few steps one has to complete before using _PFASST++_ on _JUQUEEN_.
+
+1. make sure the only modules loaded are `gcc/4.8.1` and `fftw3/3.3.3`
+
+2. prepare a custom installation of _Boost_
+
+   Because at time of writing only 1.47.0 is available compiled with _XL/C++_, which does not
+   support the C++11 features we learned to love.
+
+  1. download a recent version of _Boost_ (i.e. 1.57.0)
+
+  2. extract the archive and `cd` into it
+
+  3. run 
+
+         CXX=`which mpig++` CC=`which mpigcc` ./bootstrap.sh \
+           --with-libraries=program_options --prefix=<INSTALL_PLACE>
+
+     where `INSTALL_PLACE` is something like `$HOME/progs/juqueen`
+
+  4. create file called `user-config.jam` with the following content:
+
+         using gcc : 4.8.1 : mpig++ ;
+
+  5. now compile
+
+         CXX=`which mpig++` CC=`which mpigcc` ./b2 --prefix=<INSTALL_PLACE> \
+           --reconfigure link=static stage
+
+     and install
+
+         CXX=`which mpig++` CC=`which mpigcc` ./b2 --prefix=<INSTALL_PLACE> \
+           link=static install
+
+3. go to the sources of _PFASST++_, create a build folder and step into it
+
+4. run _CMake_
+
+       cmake -DCMAKE_TOOLCHAIN_FILE=../cmake/toolchain_juqueen.cmake \
+         -DCMAKE_CXX_COMPILER=`which mpig++` -DCMAKE_C_COMPILER=`which mpigcc` \
+         -DBOOST_ROOT=<INSTALL_PLACE> -Dpfasst_WITH_MPI=ON \
+         -Dpfasst_BUILD_TESTS=OFF ..
+
+5. run make
+
+       make
+
+6. write your _LLSubmit_ configuration file
+
+   for example for the Advection-Diffusion example:
+
+       #@job_name              = MPI_PFASST_test
+       #@comment               = "a little test of PFASST with MPI"
+       #@output                = mpi_pfasst_test_$(jobid)_$(stepid).out
+       #@error                 = mpi_pfasst_test_$(jobid)_$(stepid).err
+       #@environment           = COPY_ALL
+       #@job_type              = bluegene
+       #@notification          = never
+       #@wall_clock_limit      = 00:10:00
+       #@bg_size               = 1
+       #@bg_connectivity       = TORUS
+       #@queue
+       runjob --np 32 --ranks-per-node 32 \
+         --exe <PATH_TO_BUILD_DIR>/examples/advection_diffusion/mpi_pfasst \
+         --args '-q'