Skip to content
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.

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

[core][distributed] allow custom allreduce when pipeline parallel size > 1 #6117

Merged
merged 2 commits into from
Jul 3, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
16 changes: 5 additions & 11 deletions vllm/config.py
Original file line number Diff line number Diff line change
Expand Up @@ -723,17 +723,11 @@ def _verify_args(self) -> None:
if self.distributed_executor_backend == "ray":
from vllm.executor import ray_utils
ray_utils.assert_ray_available()
if not self.disable_custom_all_reduce and self.world_size > 1:
if is_hip():
self.disable_custom_all_reduce = True
logger.info(
"Disabled the custom all-reduce kernel because it is not "
"supported on AMD GPUs.")
elif self.pipeline_parallel_size > 1:
self.disable_custom_all_reduce = True
logger.info(
"Disabled the custom all-reduce kernel because it is not "
"supported with pipeline parallelism.")
if is_hip():
self.disable_custom_all_reduce = True
logger.info(
"Disabled the custom all-reduce kernel because it is not "
"supported on AMD GPUs.")
if self.ray_workers_use_nsight and (
not self.distributed_executor_backend == "ray"):
raise ValueError("Unable to use nsight profiling unless workers "
Expand Down
16 changes: 12 additions & 4 deletions vllm/distributed/parallel_state.py
Original file line number Diff line number Diff line change
Expand Up @@ -719,14 +719,19 @@ def init_world_group(ranks: List[int], local_rank: int,
)


def init_model_parallel_group(group_ranks: List[List[int]], local_rank: int,
backend: str) -> GroupCoordinator:
def init_model_parallel_group(
group_ranks: List[List[int]],
local_rank: int,
backend: str,
use_custom_allreduce: Optional[bool] = None) -> GroupCoordinator:
if use_custom_allreduce is None:
use_custom_allreduce = _ENABLE_CUSTOM_ALL_REDUCE
return GroupCoordinator(
group_ranks=group_ranks,
local_rank=local_rank,
torch_distributed_backend=backend,
use_pynccl=True,
use_custom_allreduce=_ENABLE_CUSTOM_ALL_REDUCE,
use_custom_allreduce=use_custom_allreduce,
)


Expand Down Expand Up @@ -888,8 +893,11 @@ def initialize_model_parallel(
for i in range(num_pipeline_model_parallel_groups):
ranks = list(range(i, world_size, num_pipeline_model_parallel_groups))
group_ranks.append(ranks)
# pipeline parallel does not need custom allreduce
_PP = init_model_parallel_group(group_ranks,
get_world_group().local_rank, backend)
get_world_group().local_rank,
backend,
use_custom_allreduce=False)


def ensure_model_parallel_initialized(
Expand Down
Loading