Merge pull request #2894 from brandonbiggs/develop

[feat] Support for `pdsh` and `clush` parallel launchers

docs/config_reference.rst

@@ -398,6 +398,7 @@ System Partition Configuration
    Available values are the following:
 
    - ``alps``: Parallel programs will be launched using the `Cray ALPS <https://pubs.cray.com/content/S-2393/CLE%205.2.UP03/cle-xc-system-administration-guide-s-2393-5203-xc/the-aprun-client>`__ ``aprun`` command.
+   - ``clush``: Parallel programs will be launched using the `ClusterShell <http://clustershell.readthedocs.org/>`__ ``clush`` command. This launcher uses the partition's :attr:`~config.systems.partitions.access` property in order to determine the options to be passed to ``clush``.
    - ``ibrun``: Parallel programs will be launched using the ``ibrun`` command.
      This is a custom parallel program launcher used at `TACC <https://portal.tacc.utexas.edu/user-guides/stampede2>`__.
    - ``local``: No parallel program launcher will be used.
@@ -406,6 +407,7 @@ System Partition Configuration
    - ``lrun-gpu``: Parallel programs will be launched using `LC Launcher <https://hpc.llnl.gov/training/tutorials/using-lcs-sierra-system#lrun>`__'s ``lrun -M "-gpu"`` command that enables the CUDA-aware Spectrum MPI.
    - ``mpirun``: Parallel programs will be launched using the ``mpirun`` command.
    - ``mpiexec``: Parallel programs will be launched using the ``mpiexec`` command.
+   - ``pdsh``: Parallel programs will be launched using the ``pdsh`` command. This launcher uses the partition's :attr:`~config.systems.partitions.access` property in order to determine the options to be passed to ``pdsh``.
    - ``srun``: Parallel programs will be launched using `Slurm <https://slurm.schedmd.com/srun.html>`__'s ``srun`` command.
    - ``srunalloc``: Parallel programs will be launched using `Slurm <https://slurm.schedmd.com/srun.html>`__'s ``srun`` command, but job allocation options will also be emitted.
      This can be useful when combined with the ``local`` job scheduler.

docs/configure.rst

@@ -116,7 +116,7 @@ Environments Configuration
 We have seen already environments to be referred to by the ``environs`` property of a partition.
 An environment in ReFrame is simply a collection of environment modules, environment variables and compiler and compiler flags definitions.
 None of these attributes is required.
-An environment can simply by empty, in which case it refers to the actual environment that ReFrame runs in.
+An environment can simply be empty, in which case it refers to the actual environment that ReFrame runs in.
 In fact, this is what the generic fallback configuration of ReFrame does.
 
 Environments in ReFrame are configured under the ``environments`` section of the documentation.

reframe/core/backends.py

@@ -13,7 +13,7 @@
 _launcher_backend_modules = [
     'reframe.core.launchers.local',
     'reframe.core.launchers.mpi',
-    'reframe.core.launchers.ssh'
+    'reframe.core.launchers.rsh'
 ]
 _launchers = {}
 _scheduler_backend_modules = [

reframe/core/launchers/ssh.py → reframe/core/launchers/rsh.py

@@ -6,6 +6,20 @@
 from reframe.core.backends import register_launcher
 from reframe.core.launchers import JobLauncher
 
+# Remote shell launchers
+
+
+@register_launcher('clush')
+class ClushLauncher(JobLauncher):
+    def command(self, job):
+        return ['clush', *job.sched_access]
+
+
+@register_launcher('pdsh')
+class PdshLauncher(JobLauncher):
+    def command(self, job):
+        return ['pdsh', *job.sched_access]
+
 
 @register_launcher('ssh')
 class SSHLauncher(JobLauncher):

unittests/test_launchers.py

@@ -11,7 +11,7 @@
 
 
 @pytest.fixture(params=[
-    'alps', 'launcherwrapper', 'local', 'mpiexec', 'mpirun',
+    'alps', 'clush', 'launcherwrapper', 'local', 'mpiexec', 'mpirun', 'pdsh',
     'srun', 'srunalloc', 'ssh', 'upcrun', 'upcxx-run', 'lrun', 'lrun-gpu'
 ])
 def launcher(request):
@@ -66,8 +66,11 @@ def _make_job(launcher, *args, **kwargs):
 
 @pytest.fixture()
 def job(make_job, launcher):
-    if type(launcher).registered_name == 'ssh':
+    launcher_name = type(launcher).registered_name
+    if launcher_name == 'ssh':
         access = ['-l user', '-p 22222', 'host']
+    elif launcher_name in ('clush', 'pdsh'):
+        access = ['-w', 'hostA', '-x', 'hostB']
     else:
         access = None
 
@@ -92,8 +95,11 @@ def job(make_job, launcher):
 
 @pytest.fixture
 def minimal_job(make_job, launcher):
-    if type(launcher).registered_name == 'ssh':
+    launcher_name = type(launcher).registered_name
+    if launcher_name == 'ssh':
         access = ['host']
+    elif launcher_name in ('clush', 'pdsh'):
+        access = ['-w', 'host']
     else:
         access = None
 
@@ -138,6 +144,8 @@ def test_run_command(job):
                            '--foo')
     elif launcher_name == 'ssh':
         assert command == 'ssh -o BatchMode=yes -l user -p 22222 --foo host'
+    elif launcher_name in ('clush', 'pdsh'):
+        assert command == f'{launcher_name} -w hostA -x hostB --foo'
     elif launcher_name == 'upcrun':
         assert command == 'upcrun -N 2 -n 4 --foo'
     elif launcher_name == 'upcxx-run':
@@ -175,6 +183,8 @@ def test_run_command_minimal(minimal_job):
                            '--foo')
     elif launcher_name == 'ssh':
         assert command == 'ssh -o BatchMode=yes --foo host'
+    elif launcher_name in ('clush', 'pdsh'):
+        assert command == f'{launcher_name} -w host --foo'
     elif launcher_name == 'upcrun':
         assert command == 'upcrun -n 1 --foo'
     elif launcher_name == 'upcxx-run':