Skip to content

[None][chore] Fix slurm job name #11265

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
yingguo-trt merged 6 commits into from
Feb 5, 2026
Merged

[None][chore] Fix slurm job name #11265

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e6d0362
fix: Update SLURM job names to follow cluster convention
yingguo-trt Feb 4, 2026
00eeb5a
fix: Update SLURM job names to follow cluster convention
yingguo-trt Feb 4, 2026
ee49aa2
fix pre-commit error
yingguo-trt Feb 4, 2026
375f0c5
fix: Update SLURM job names to follow cluster convention
yingguo-trt Feb 4, 2026
c252b92
use SLURM_ACCOUNT as job name prefix instead of hardcoding
yingguo-trt Feb 5, 2026
1571d82
fix pre-commit error
yingguo-trt Feb 5, 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
1 change: 1 addition & 0 deletions tests/integration/defs/perf/disagg/execution/executor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -219,6 +219,7 @@ def submit_test_job(test_config) -> tuple:

# Write temporary config file with replaced environment variables
logger.info(f"Creating temporary config: {temp_config_path}")

with open(temp_config_path, "w") as f:
yaml.dump(
test_config.config_data,
Expand Down
54 changes: 29 additions & 25 deletions tests/integration/defs/perf/disagg/utils/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,49 +2,42 @@

import os

# GPU resource configuration
# Centralized configuration for all GPU-specific parameters
# GPU resource configuration - centralized config for all GPU-specific parameters
GPU_RESOURCE_CONFIG = {
# OCI GB200
"GB200": {
"slurm_extra_args": "--gres=gpu:4", # SLURM extra arguments (empty string if not required)
"GB200": { # OCI GB200
"slurm_extra_args": "--gres=gpu:4",
"set_segment": True,
"lock_freq_graphics_mhz": 2062, # GPU graphics clock lock frequency (MHz)
"lock_freq_memory_mhz": 3996, # GPU memory clock lock frequency (MHz)
"lock_freq_graphics_mhz": 2062,
"lock_freq_memory_mhz": 3996,
},
# Lyris GB200
"GB200_LYRIS": {
"slurm_extra_args": "", # GB300 does not require extra args
"GB200_LYRIS": { # Lyris GB200
"slurm_extra_args": "",
"set_segment": True,
"lock_freq_graphics_mhz": None, # TODO: Set GB300 lock frequency
"lock_freq_graphics_mhz": None,
"lock_freq_memory_mhz": None,
},
# Lyris GB300
"GB300": {
"slurm_extra_args": "", # GB300 does not require extra args
"GB300": { # Lyris GB300
"slurm_extra_args": "",
"set_segment": True,
"lock_freq_graphics_mhz": None, # TODO: Set GB300 lock frequency
"lock_freq_graphics_mhz": None,
"lock_freq_memory_mhz": None,
},
# H100
"H100": {
"slurm_extra_args": "", # H100 does not require extra args
"slurm_extra_args": "",
"set_segment": False,
"lock_freq_graphics_mhz": None, # TODO: Set H100 lock frequency
"lock_freq_graphics_mhz": None,
"lock_freq_memory_mhz": None,
},
# B200
"B200": {
"B200": { # OCI B200
"slurm_extra_args": "--gres=gpu:4",
"set_segment": False,
"lock_freq_graphics_mhz": None, # TODO: Set B200 lock frequency
"lock_freq_graphics_mhz": None,
"lock_freq_memory_mhz": None,
},
# B300
"B300": {
"B300": { # OCI B300
"slurm_extra_args": "--gres=gpu:4",
"set_segment": False,
"lock_freq_graphics_mhz": None, # TODO: Set B300 lock frequency
"lock_freq_graphics_mhz": None,
"lock_freq_memory_mhz": None,
},
}
Expand All @@ -67,7 +60,18 @@ def get_slurm_account() -> str:

@staticmethod
def get_slurm_job_name() -> str:
return os.getenv("SLURM_JOB_NAME", "unified-benchmark")
"""Get SLURM job name: {SLURM_ACCOUNT}-{base}.

Example: myaccount-unified.benchmark
Customize base via SLURM_JOB_BASE_NAME env var (default: unified.benchmark)
"""
account = EnvManager.get_slurm_account()
base = os.getenv("SLURM_JOB_BASE_NAME", "unified.benchmark")

# Only use account as prefix if it's set and not a placeholder
if account and not account.startswith("<"):
return f"{account}-{base}"
return base

@staticmethod
def get_slurm_set_segment() -> bool:
Expand Down