Skip to content

fix: Pass instance type into aws batch job definition even when num_replicas = 1 #1102

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
merged 1 commit into from
Sep 10, 2025
Merged

fix: Pass instance type into aws batch job definition even when num_replicas = 1 #1102

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
2 changes: 1 addition & 1 deletion torchx/schedulers/aws_batch_scheduler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -255,7 +255,7 @@ def _role_to_node_properties(
container["jobRoleArn"] = job_role_arn
if execution_role_arn:
container["executionRoleArn"] = execution_role_arn
if role.num_replicas > 1:
if role.num_replicas > 0:
instance_type = instance_type_from_resource(role.resource)
if instance_type is not None:
container["instanceType"] = instance_type
Expand Down
4 changes: 2 additions & 2 deletions torchx/schedulers/test/aws_batch_scheduler_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -195,15 +195,15 @@ def test_submit_dryrun_instance_type_multinode(self) -> None:
node_groups[0]["container"]["instanceType"],
)

def test_submit_dryrun_no_instance_type_singlenode(self) -> None:
def test_submit_dryrun_instance_type_singlenode(self) -> None:
cfg = AWSBatchOpts({"queue": "ignored_in_test", "privileged": True})
resource = specs.named_resources_aws.aws_p3dn_24xlarge()
app = _test_app(num_replicas=1, resource=resource)
info = create_scheduler("test").submit_dryrun(app, cfg)
# pyre-ignore[16]
node_groups = info.request.job_def["nodeProperties"]["nodeRangeProperties"]
self.assertEqual(1, len(node_groups))
self.assertTrue("instanceType" not in node_groups[0]["container"])
self.assertTrue("instanceType" in node_groups[0]["container"])

def test_submit_dryrun_no_instance_type_non_aws(self) -> None:
cfg = AWSBatchOpts({"queue": "ignored_in_test", "privileged": True})
Expand Down
Loading