Skip to content
Snippets Groups Projects
Select Git revision
  • ba9bb9af4406b1228b7a34509fdc9e3dba111be9
  • devel default
  • 107-compilation-error-when-building-maestro-core-on-m1-apple-processors
  • 108-implement-cpu-id-query-for-apple-m1-hardware
  • 58-scripting-interface-to-maestro-core
  • 101-need-ci-test-using-installed-maestro
  • 57-sphinx-documentation
  • 105-memory-leak-in-pm-message-envelope-handling
  • 104-permit-disabling-memory-pool
  • 103-liberl-installation-issue-on-devel
  • 94-maestro-rdma-transport-ignores-max_msg_size-2
  • main protected
  • 102-possible-race-in-check_pm_redundant_interlock-test
  • 97-check-if-shm-provider-can-be-enabled-after-libfabric-1-14-is-in-our-tree-2
  • 100-include-maestro-attributes-h-cannot-include-mamba-header-from-deps-path
  • 97-check-if-shm-provider-can-be-enabled-after-libfabric-1-14-is-in-our-tree
  • 17-job-failed-282354-needs-update-of-mio-interface-and-build-rules
  • 96-test-libfabric-update-to-1-13-or-1-14
  • feature/stop-telemetry-after-all-left
  • 94-maestro-rdma-transport-ignores-max_msg_size
  • 93-improve-performance-of-mstro_attribute_val_cmp_str
  • v0.3_rc1
  • maestro_d65
  • d65_experiments_20211113
  • v0.2
  • v0.2_rc1
  • d3.3
  • d3.3-review
  • d5.5
  • d5.5-review
  • v0.1
  • d3.2
  • d3.2-draft
  • v0.0
34 results

build-envs

A Load Balancing Library (ALL)

The library aims to provide an easy way to include dynamic domain-based load balancing into particle based simulation codes. The library is developed in the Simulation Laboratory Molecular Systems of the Jülich Supercomputing Centre at Forschungszentrum Jülich.

Only a brief summary is given here and more information can be found in the official documentation such as a detailed API description, examples and further information regarding the load balancing methods.

Installation and Requirements

Requirements

Base requirements:

  • C++11 capable compiler
  • MPI support
  • CMake v. 3.14 or higher

Optional requirements:

  • Fortran 2003 capable compiler (Fortran interface)
  • Fortran 2008 capable compiler (Usage of the mpi_f08 interface)
  • VTK 7.1 or higher (Domain output)
  • Boost testing utilities
  • Doxygen and Sphinx with breathe (Documentation)

Installation

  1. Clone the library from https://gitlab.version.fz-juelich.de/SLMS/loadbalancing into $ALL_ROOT_DIR.
  2. Create the build directory $ALL_BUILD_DIR some place else.
  3. Call cmake -S "$ALL_ROOT_DIR" -B "$ALL_BUILD_DIR" to set up the installation. To use a specific compiler and Boost installation use: CC=gcc CXX=g++ BOOST_ROOT=$BOOST_DIR cmake [...].
  4. To build and install the library then run: cmake --build "$ALL_BUILD_DIR" and cmake --install "$ALL_BUILD_DIR" --prefix "$ALL_INSTALL_DIR". Afterwards, the built examples and library files are placed in $ALL_INSTALL_DIR.