Spack is a package manager for supercomputers, Linux, and macOS. It makes installing scientific software easy. With Spack, you can build a package with multiple versions, configurations, platforms, and compilers, and all of these builds can coexist on the same machine.

Spack isn’t tied to a particular language; you can build a software stack in Python or R, link to libraries written in C, C++, or Fortran, easily swap compilers, and target specific microarchitectures. Use Spack to install without root in your home directory, to manage shared installations and modules, or to build combinatorial versions of software for testing.

Install Spack

Clone Spack from GitHub and you’re ready to go:

git clone https://github.com/spack/spack.git
. spack/share/spack/setup-env.sh
spack install hdf5

Custom versions & configurations

The installation of Spack can be customized in a variety of ways. Users can specify the package version, compiler, compile-time options, and even cross-compile platform, all from the command line.

   # Install a specific version by appending @
   $ spack install hdf5@1.10.1

   # Specify a compiler (and optional version), with %
   $ spack install hdf5@1.10.1 %gcc@4.7.3

   # Add special boolean compile-time options with +
   $ spack install hdf5@1.10.1 %gcc@4.7.3 +szip

   # Add custom compiler flags
   $ spack install hdf5@1.10.1 %gcc@4.7.3 cppflags="-O3 -floop-block"

   # Cross-compile for compute nodes on a Cray or Blue Gene/Q
   $ spack install hdf5@1.10.1 target=backend

Users can specify as many or few options as they care about. Spack will fill in the unspecified values with sensible defaults. The two listed syntaxes for variants are identical when the value is boolean.

Customize dependencies

Spack allows dependencies of particular installations to be customized extensively. Suppose that hdf5 depends on openmpi and indirectly on hwloc. Using ^, users can add custom configurations for dependencies:

   # Install hdf5 and link it with specific versions of openmpi and hwloc
   $ spack install hdf5@1.10.1 %gcc@4.7.3 +debug ^openmpi+cuda fabrics=auto ^hwloc+gl

Packages can peacefully coexist

Spack installs every unique package/dependency configuration into its own prefix, so new installs will not break existing ones.

Spack avoids library misconfiguration by using RPATH to link dependencies. When a user links a library or runs a program, it is tied to the dependencies it was built with, so there is no need to manipulate LD_LIBRARY_PATH at runtime.

Creating packages is easy

Spack packages are simple Python scripts. The spack create command will generate boilerplate to get you started, and you can create a package in a matter of minutes. You write the build instructions; Spack builds the dependencies for you.

from spack import *

class Kripke(Package):
    """Kripke is a simple, scalable, 3D Sn deterministic particle
       transport proxy/mini app.
    """
    homepage = "https://computing.llnl.gov/projects/co-design/kripke"
    url      = "https://computing.llnl.gov/downloads/kripke-openmp-1.1.tar.gz"

    version('1.1', '7fe6f2b26ed983a6ce5495ab701f85bf')

    variant('mpi',    default=True, description='Build with MPI.')
    variant('openmp', default=True, description='Build with OpenMP enabled.')

    depends_on('mpi', when="+mpi")

    def install(self, spec, prefix):
        with working_dir('build', create=True):
            cmake('-DCMAKE_INSTALL_PREFIX:PATH=.',
                  '-DENABLE_OPENMP=%s' % ('+openmp' in spec),
                  '-DENABLE_MPI=%s'    % ('+mpi' in spec),
                  '..',
                  *std_cmake_args)

            make()
            mkdirp(prefix.bin)
            install('kripke', prefix.bin)

Get Involved!

Visit Spack on GitHub and take the tutorial. Join the discussion on the GoogleGroup, and learn how to contribute your own packages. Check out the Upcoming Events page for tutorials, workshops, BoFs, etc.