mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-02-16 15:55:08 +08:00
[None][feat] support Lyris GB200 and increase disagg test timeout (#11019)
Signed-off-by: yingguo-trt <244492186+yingguo-trt@users.noreply.github.com> Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com> Co-authored-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com> Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
This commit is contained in:
parent
fd4e6132e5
commit
d348dd95a7
@ -13,7 +13,7 @@ slurm:
|
||||
script_file: disaggr_torch.slurm
|
||||
partition: <partition>
|
||||
account: <account>
|
||||
job_time: 03:00:00
|
||||
job_time: 04:00:00
|
||||
job_name: unified-benchmark
|
||||
extra_args: "--gres=gpu:4"
|
||||
numa_bind: true
|
||||
|
||||
@ -14,9 +14,9 @@ GPU_RESOURCE_CONFIG = {
|
||||
},
|
||||
# Lyris GB200
|
||||
"GB200_LYRIS": {
|
||||
"slurm_extra_args": "", # GB200 does not require extra args
|
||||
"slurm_extra_args": "", # GB300 does not require extra args
|
||||
"set_segment": True,
|
||||
"lock_freq_graphics_mhz": None, # TODO: Set GB200 lock frequency
|
||||
"lock_freq_graphics_mhz": None, # TODO: Set GB300 lock frequency
|
||||
"lock_freq_memory_mhz": None,
|
||||
},
|
||||
# Lyris GB300
|
||||
|
||||
@ -343,7 +343,6 @@ class ConfigLoader:
|
||||
metadata = config_data.get("metadata", {})
|
||||
model_name = metadata.get("model_name", "unknown")
|
||||
supported_gpus = metadata.get("supported_gpus", ["GB200", "GB300", "H100", "B200", "B300"])
|
||||
|
||||
# Override config with environment variables (in memory only, do not write back)
|
||||
config_data = self._apply_env_overrides(config_data, model_name)
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user