biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Fixes and Optimizations for DeepEP + DeepGEMM combination. (#19298)

Browse Source 
Signed-off-by: Varun <vsundarr@redhat.com>
Co-authored-by: Varun <vsundarr@redhat.com>
This commit is contained in:
Varun Sundar Rabindranath
2025-06-09 10:50:39 -04:00
committed by GitHub
parent b8089195b4
commit 5cf2daea9a
8 changed files with 98 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
vllm/distributed/device_communicators/all2all.py
View File

@@ -233,16 +233,11 @@ class DeepEPLLAll2AllManager(DeepEPAll2AllManagerBase):
# Defaults for internode and intranode are taken from DeepEP tests.
num_nvl_bytes = 1024 * 1024 * 1024
num_qps_per_rank = num_local_experts
num_rdma_bytes = None
if self.internode:
num_rdma_bytes = 1024 * 1024 * 1024
else:
num_rdma_bytes = deep_ep.Buffer.get_low_latency_rdma_size_hint(
num_max_dispatch_tokens_per_rank=max_num_tokens_per_dp_rank,
hidden=token_hidden_size,
num_ranks=num_ep_ranks,
num_experts=num_global_experts)
num_rdma_bytes = deep_ep.Buffer.get_low_latency_rdma_size_hint(
num_max_dispatch_tokens_per_rank=max_num_tokens_per_dp_rank,
hidden=token_hidden_size,
num_ranks=num_ep_ranks,
num_experts=num_global_experts)
assert num_rdma_bytes is not None
return dict(group=self.cpu_group,