[EPLB] Add nixl-based eplb communicator (#36276)

Signed-off-by: ilmarkov <markovilya197@gmail.com>
Signed-off-by: Markov Ilya <markovilya19@gmail.com>
This commit is contained in:
Ilya Markov
2026-04-20 10:24:23 +00:00
committed by GitHub
parent f774ba028a
commit 50dd4cb427
10 changed files with 556 additions and 80 deletions
+20 -14
View File
@@ -541,23 +541,29 @@ def rearrange_expert_weights_inplace(
assert num_physical_experts == ep_size * num_local_physical_experts
first_layer_weights = list(expert_weights[0])
if is_profile:
if communicator.needs_profile_buffer_reservation:
# Reserve NCCL communication buffers via a dummy all_gather.
# Backends that pre-allocate their own transfer buffers
# skip this to avoid the extra memory spike during profiling.
weights_buffer: list[torch.Tensor] = [
torch.empty_like(w) for w in first_layer_weights
]
for weight, buffer in zip(expert_weights[0], weights_buffer):
dummy_recv_buffer = [buffer for _ in range(ep_size)]
torch.distributed.barrier()
all_gather(
dummy_recv_buffer,
weight,
group=ep_group,
)
return
# Buffers to hold the expert weights during the exchange.
# NOTE: Currently we assume the same weights across different layers
# have the same shape.
weights_buffer: list[torch.Tensor] = [
torch.empty_like(w) for w in first_layer_weights
]
if is_profile:
# Reserve communication buffers via a minimal dummy all_gather on first layer
for weight, buffer in zip(expert_weights[0], weights_buffer):
dummy_recv_buffer = [buffer for _ in range(ep_size)]
torch.distributed.barrier()
all_gather(
dummy_recv_buffer,
weight,
group=ep_group,
)
return
weights_buffer = [torch.empty_like(w) for w in first_layer_weights]
# NOTE(bowen): We need this synchronize to run, but I don't know why.
# If you figure out the reason, please let me know -- thank you!