Skip to content

Apply the capped memory overhead to a job that requests exactly the cap #1007

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

Open
calvinp0 wants to merge 1 commit into
base: main
Choose a base branch
from
+21 −3
Open

Apply the capped memory overhead to a job that requests exactly the cap #1007

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions arc/job/adapter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -591,9 +591,9 @@ def set_cpu_and_mem(self):
self.cpu_cores = self.cpu_cores or job_cpu_cores
max_mem = servers[self.server].get('memory', None) if self.server is not None else 32.0 # Max memory per node in GB.
job_max_server_node_memory_allocation = default_job_settings.get('job_max_server_node_memory_allocation', 0.95)
if max_mem is not None and self.job_memory_gb > max_mem * job_max_server_node_memory_allocation:
if max_mem is not None and self.job_memory_gb >= max_mem * job_max_server_node_memory_allocation:
logger.warning(f'The memory for job {self.job_name} using {self.job_adapter} ({self.job_memory_gb} GB) '
f'exceeds {100 * job_max_server_node_memory_allocation}% of the the maximum node memory on '
f'reaches {100 * job_max_server_node_memory_allocation}% of the the maximum node memory on '
f'{self.server}. Setting it to {job_max_server_node_memory_allocation * max_mem:.2f} GB.')
self.job_memory_gb = job_max_server_node_memory_allocation * max_mem
total_submit_script_memory_mib = math.ceil(self.job_memory_gb * MEMORY_GB_TO_MIB * CAPPED_JOB_MEMORY_OVERHEAD)
Expand Down
20 changes: 19 additions & 1 deletion arc/job/adapter_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,8 @@
from arc.level import Level
from arc.species import ARCSpecies

servers, submit_filenames = settings['servers'], settings['submit_filenames']
default_job_settings, servers, submit_filenames = \
settings['default_job_settings'], settings['servers'], settings['submit_filenames']


class TestEnumerationClasses(unittest.TestCase):
Expand Down Expand Up @@ -250,6 +251,23 @@ def test_set_cpu_and_mem(self):
self.assertEqual(self.job_4.submit_script_memory, expected_memory)
self.job_4.server = 'local'

capped_memory_gb = servers['server2']['memory'] \
* default_job_settings['job_max_server_node_memory_allocation']
original_job_memory_gb = self.job_4.job_memory_gb
self.addCleanup(setattr, self.job_4, 'job_memory_gb', original_job_memory_gb)
self.addCleanup(setattr, self.job_4, 'server', 'local')
self.job_4.server = 'server2'
self.job_4.cpu_cores = None
self.job_4.job_memory_gb = capped_memory_gb
self.job_4.set_cpu_and_mem()
self.assertEqual(self.job_4.job_memory_gb, capped_memory_gb)
self.assertEqual(self.job_4.submit_script_memory_mib, math.ceil(capped_memory_gb * 1024 * 1.05))
self.assertLess(self.job_4.submit_script_memory_mib, servers['server2']['memory'] * 1024)
self.assertIn('max_total_job_memory', self.job_4.job_status[1]['keywords'])
self.job_4.job_status[1]['keywords'].remove('max_total_job_memory')
self.job_4.job_memory_gb = original_job_memory_gb
self.job_4.server = 'local'

def test_set_file_paths(self):
"""Test setting up the job's paths"""
self.assertEqual(self.job_1.local_path, os.path.join(self.job_1.project_directory, 'calcs', 'Species',
Expand Down
Loading