Skip to content

fix ray benchmark #1016

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
anson627 merged 7 commits into from
Jan 14, 2026
Merged

fix ray benchmark #1016

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a2a7195
fix typo
anson627 Jan 14, 2026
19472ab
fix job template
anson627 Jan 14, 2026
4fb03a3
fix job template path
anson627 Jan 14, 2026
775026f
fix chart version
anson627 Jan 14, 2026
22c8425
fix cl2 config dir
anson627 Jan 14, 2026
80f8f6c
update job template path
anson627 Jan 14, 2026
d304758
Merge branch 'main' into fix-ray-benchmark
anson627 Jan 14, 2026
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 modules/python/clusterloader2/job_controller/config/ray/config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,7 @@ steps:
Params:
action: start
labelSelector: app.kubernetes.io/created-by = kuberay-operator
measurmentInterval: 1s
measurementInterval: 1s

- name: Creating RayJobs for PyTorch MNIST fine-tuning
phases:
Expand Down
10 changes: 5 additions & 5 deletions modules/python/clusterloader2/job_controller/config/ray/job_template.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ spec:
effect: "NoSchedule"
containers:
- name: ray-submitter
image: rayproject/ray:2.41.0
image: {{.Image}}
rayClusterSpec:
rayVersion: '2.9.3'
headGroupSpec:
Expand All @@ -38,7 +38,7 @@ spec:
effect: "NoSchedule"
containers:
- name: ray-head
image: rayproject/ray:2.41.0
image: {{.Image}}
ports:
- containerPort: 6379
name: gcs-server
Expand All @@ -65,10 +65,10 @@ spec:
effect: "NoSchedule"
containers:
- name: ray-worker
image: rayproject/ray:2.41.0
image: {{.Image}}
resources:
limits:
nvidia.com/gpu: 1
nvidia.com/gpu: {{.JobGPU}}
requests:
nvidia.com/gpu: 1
nvidia.com/gpu: {{.JobGPU}}

8 changes: 7 additions & 1 deletion modules/python/clusterloader2/job_controller/job_controller.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -152,7 +152,8 @@ def install_ray_dependencies(self):
from "<cl2_config_dir>/ray/".
- Waits for operator and mock-head pods to be ready.
"""
config_dir = os.path.join("./clusterloader2/job_controller/config", "ray")
logger.info("cl2 config dir: %s", self.cl2_config_dir)
config_dir = os.path.join(self.cl2_config_dir, "ray")
values_file = os.path.join(config_dir, "values.yaml")

# Install KubeRay operator via Helm
Expand Down Expand Up @@ -180,6 +181,8 @@ def install_ray_dependencies(self):
"--install",
"kuberay-operator",
"kuberay/kuberay-operator",
"--version",
"1.4.2",
"--namespace",
"kuberay-system",
"--create-namespace",
Expand Down Expand Up @@ -283,6 +286,9 @@ def add_configure_subparser_arguments(parser):
type=str,
help="Timeout before failing the scale up test",
)
parser.add_argument(
"--cl2_config_dir", type=str, help="Path to the CL2 config directory"
)
parser.add_argument(
"--cl2_override_file",
type=str,
Expand Down
1 change: 1 addition & 0 deletions steps/engine/clusterloader2/job_controller/execute.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ steps:
--dra_enabled ${ENABLE_DRA:-False} \
--ray_enabled ${ENABLE_RAY:-False} \
--job_gpu ${JOB_GPU:-0} \
--cl2_config_dir ${CL2_CONFIG_DIR} \
--cl2_override_file ${CL2_CONFIG_DIR}/overrides.yaml

PYTHONPATH=$PYTHONPATH:$(pwd) python3 $PYTHON_SCRIPT_FILE execute \
Expand Down