[EPLB] Nixl communicator optimization. Zero-copy transfers (#41633)

Signed-off-by: ilmarkov <markovilya197@gmail.com>
Signed-off-by: Markov Ilya <markovilya19@gmail.com>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Markov Ilya <markovilya19@gmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
This commit is contained in:
Ilya Markov
2026-06-04 03:40:34 +00:00
committed by GitHub
co-authored by Markov Ilya Nicolò Lucchesi Tyler Michael Smith Tyler Michael Smith
parent f0cd590d62
commit 4f423bd5bc
12 changed files with 287 additions and 219 deletions
@@ -40,11 +40,7 @@ def _can_p2p(rank: int, world_size: int) -> bool:
return True
def is_weak_contiguous(inp: torch.Tensor):
return inp.is_contiguous() or (
inp.storage().nbytes() - inp.storage_offset() * inp.element_size()
== inp.numel() * inp.element_size()
)
from vllm.distributed.utils import is_weak_contiguous # noqa: E402
class CustomAllreduce:
@@ -24,11 +24,7 @@ except Exception:
quick_ar = False
def is_weak_contiguous(inp: torch.Tensor):
return inp.is_contiguous() or (
inp.storage().nbytes() - inp.storage_offset() * inp.element_size()
== inp.numel() * inp.element_size()
)
from vllm.distributed.utils import is_weak_contiguous # noqa: E402, F401
class QuickReduceRegime(Enum):