Skip to content

Resolve failing to submit to hortense GPU partition #250

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
casparvl merged 8 commits into from
Apr 24, 2025
Merged

Resolve failing to submit to hortense GPU partition #250

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
042fa51
Resolve failing to submit to hortense GPU partition
laraPPr Mar 4, 2025
52b0030
add reference to issue in config
laraPPr Mar 4, 2025
6f40c19
Update vsc_hortense.py
laraPPr Mar 4, 2025
96e7615
update the ci_config for running EESSI on Hortense
Mar 10, 2025
291c136
use env_vars instead of not purging the environment
Apr 17, 2025
d537c42
make linter happy
Apr 17, 2025
5804a7a
fix comment
Apr 17, 2025
8cb80e5
Merge branch 'EESSI:main' into CI_Hortense
laraPPr Apr 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 14 additions & 12 deletions CI/hortense_EESSI_ss/ci_config.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,19 +1,21 @@
# Configurable items
if [ -z "${TEST_SUITE_PARTITION}" ]; then
echo "You have to indicate on which partition the test-suite will run on vsc-Hortense"
echo "This environment variable needs to be set TEST_SUITE_PARTITION=cpu_rome_256gb"
echo "Can only set to 'cpu_rome_256gb' untill new functionality of 'sched_options' is part of"
echo "the ReFrame release https://github.com/reframe-hpc/reframe/issues/2970"
exit 1
if [ -z "${UNSET_MODULEPATH}" ]; then
export UNSET_MODULEPATH=False
module --force purge
fi

if [ -z "${REFRAME_ARGS}" ]; then
REFRAME_ARGS="--tag CI --tag 1_core --system hortense:${TEST_SUITE_PARTITION}"
if [[ "$TEST_SUITE_PARTITION" == "GPU" ]]; then
if [ -z "${SET_LOCAL_MODULE_ENV}"]; then
export SET_LOCAL_MODULE_ENV=True
fi
if [ -z "${LOCAL_MODULES}"]; then
export LOCAL_MODULES="env/slurm/dodrio/gpu_rome_a100"
module use /cvmfs/software.eessi.io/versions/2023.06/software/linux/x86_64/amd/zen2/accel/nvidia/cc80/modules/all
fi
fi

if [ -z "${UNSET_MODULEPATH}" ]; then
export UNSET_MODULEPATH=False
module --force purge
if [ -z "${REFRAME_ARGS}" ]; then
REFRAME_ARGS="--tag CI --tag 1_core"
fi

if [ -z "${USE_EESSI_SOFTWARE_STACK}" ]; then
Expand All @@ -29,5 +31,5 @@ if [ -z "${SET_LOCAL_MODULE_ENV}"]; then
fi

if [ -z "${LOCAL_MODULES}"]; then
export LOCAL_MODULES="env/vsc/dodrio/${TEST_SUITE_PARTITION} env/slurm/dodrio/${TEST_SUITE_PARTITION}"
export LOCAL_MODULES="env/slurm/dodrio/cpu_rome"
fi
13 changes: 11 additions & 2 deletions config/vsc_hortense.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,6 +57,9 @@ def command(self, job):
'descr': 'Hortense',
'hostnames': ['login.*.dodrio.os'],
'modules_system': 'lmod',
# Need to set the environment variable to be able to submit to GPU_partitions
# see https://github.com/EESSI/test-suite/issues/242
'env_vars': [['SLURM_CONF', '/etc/slurm/slurm.conf_dodrio']],
'partitions': [
{
'name': 'cpu_rome',
Expand Down Expand Up @@ -145,7 +148,10 @@ def command(self, job):
{
'name': 'gpu_rome_a100_40',
'scheduler': 'slurm',
'prepare_cmds': [prepare_eessi_init, common_eessi_init()],
'prepare_cmds': [
prepare_eessi_init,
common_eessi_init(),
],
'access': hortense_access + ['--partition=gpu_rome_a100_40'],
'sched_options': {
'sched_access_in_submit': True,
Expand Down Expand Up @@ -185,7 +191,10 @@ def command(self, job):
{
'name': 'gpu_rome_a100_80',
'scheduler': 'slurm',
'prepare_cmds': [prepare_eessi_init, common_eessi_init()],
'prepare_cmds': [
prepare_eessi_init,
common_eessi_init(),
],
'access': hortense_access + ['--partition=gpu_rome_a100_80'],
'sched_options': {
'sched_access_in_submit': True,
Expand Down