[RL] Add sharding-aware NCCL M2N weight transfer (#51520)
Signed-off-by: Ke Wen <kwen@nvidia.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com> Co-authored-by: aoshen02 <aoshen@inferact.ai>
This commit is contained in:
co-authored by
mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
aoshen02
parent
4bb804cc5b
commit
b6d2b4fc25
@@ -220,11 +220,11 @@ def worker_init_process_group(
|
||||
|
||||
def trainer_init(
|
||||
init_info: NCCLRendezvous | dict,
|
||||
rank: int = 0,
|
||||
) -> "PyNcclCommunicator":
|
||||
"""Initialize NCCL process group for trainer-side weight transfer.
|
||||
|
||||
The trainer is always rank 0 in the process group. Uses the current
|
||||
CUDA device (torch.accelerator.current_device_index()).
|
||||
Uses the current CUDA device (torch.accelerator.current_device_index()).
|
||||
|
||||
Args:
|
||||
init_info: Any object carrying the `NCCLRendezvous` fields (a trainer or
|
||||
@@ -232,6 +232,9 @@ def trainer_init(
|
||||
- master_address: str
|
||||
- master_port: int
|
||||
- world_size: int
|
||||
rank: This trainer process's rank in the group. The broadcast backends
|
||||
have a single trainer at rank 0; multi-rank trainers (m2n) occupy
|
||||
ranks `[0, num_trainer_ranks)` and the workers start after them.
|
||||
|
||||
Returns:
|
||||
PyNcclCommunicator for weight transfer.
|
||||
@@ -246,12 +249,11 @@ def trainer_init(
|
||||
master_port = init_info.master_port
|
||||
world_size = init_info.world_size
|
||||
|
||||
# Trainer is always rank 0
|
||||
device = torch.accelerator.current_device_index()
|
||||
return stateless_init_process_group(
|
||||
master_address,
|
||||
master_port,
|
||||
0,
|
||||
rank,
|
||||
world_size,
|
||||
device,
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user