[Docs] Fix NCCL typo

jiangkuaixue123 · jiangkuaixue123 · commit 9ddb48cc9f8d · 2026-02-14T11:38:55.000+08:00
Signed-off-by: jiangkuaixue123 &lt;jiangxiaozhou111@163.com&gt;
diff --git a/docs/design/p2p_nccl_connector.md b/docs/design/p2p_nccl_connector.md
@@ -52,7 +52,7 @@ Currently, only symmetric TP (Tensor Parallelism) methods are supported for KVCa
 
 ![image2](https://github.com/user-attachments/assets/837e61d6-365e-4cbf-8640-6dd7ab295b36)
 
-Each NCCL group occupies a certain amount of GPU memory buffer for communication, the size of which is primarily influenced by the `NCCL_MAX_NCHANNELS` environment variable. When `NCCL_MAX_NCHANNELS=16`, an NCCL group typically occupies 100MB, while when `NCCL_MAX_NCHANNELS=8`, it usually takes up 52MB. For large-scale xPyD configurations—such as DeepSeek's 96P144D—this implementation is currently not feasible. Moving forward, we are considering using RDMA for point-to-point communication and are also keeping an eye on UCCL.
+Each NCCL group occupies a certain amount of GPU memory buffer for communication, the size of which is primarily influenced by the `NCCL_MAX_NCHANNELS` environment variable. When `NCCL_MAX_NCHANNELS=16`, an NCCL group typically occupies 100MB, while when `NCCL_MAX_NCHANNELS=8`, it usually takes up 52MB. For large-scale xPyD configurations—such as DeepSeek's 96P144D—this implementation is currently not feasible. Moving forward, we are considering using RDMA for point-to-point communication and are also keeping an eye on NCCL.
 
 ### GPU Memory Buffer and Tensor Memory Pool
 
