EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[PD] Make bootstrap code common between NIXL and Mooncake (#6473)

Browse Source
This commit is contained in:
Trevor Morris
2025-05-27 12:47:38 -07:00
committed by GitHub
parent fa6723f08f
commit e806f708c9
6 changed files with 596 additions and 429 deletions
Download Patch File Download Diff File Expand all files Collapse all files

41
docs/backend/pd_disaggregation.md
View File

@@ -47,3 +47,44 @@ $ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 --dis
# decode 1
$ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 --disaggregation-ib-device ${device_name} --disaggregation-mode decode --host ${local_ip} --port 30001 --trust-remote-code --dist-init-addr ${decode_master_ip}:5000 --nnodes 2 --node-rank 1 --tp-size 16 --dp-size 8 --enable-dp-attention --enable-deepep-moe --deepep-mode low_latency --mem-fraction-static 0.8 --max-running-requests 128
```
## NIXL
### Requirements
Install via pip.
```bash
pip install nixl
```
Or build from source - may be required if you already have UCX installed.
```bash
git clone https://github.com/ai-dynamo/nixl.git
cd nixl
pip install . --config-settings=setup-args="-Ducx_path=/path/to/ucx"
```
### Usage
### Llama Single Node
```bash
$ python -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --disaggregation-mode prefill --disaggregation-transfer-backend nixl
$ python -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --disaggregation-mode decode --port 30001 --base-gpu-id 1 --disaggregation-transfer-backend nixl
$ python -m sglang.srt.disaggregation.mini_lb --prefill http://127.0.0.1:30000 --decode http://127.0.0.1:30001 --host 0.0.0.0 --port 8000
```
### DeepSeek Multi-Node
```bash
# prefill 0
$ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 ---disaggregation-transfer-backend nixl --disaggregation-mode prefill --host ${local_ip} --port 30000 --trust-remote-code --dist-init-addr ${prefill_master_ip}:5000 --nnodes 2 --node-rank 0 --tp-size 16 --dp-size 8 --enable-dp-attention --enable-deepep-moe --deepep-mode normal --mem-fraction-static 0.8
# prefill 1
$ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 ---disaggregation-transfer-backend nixl --disaggregation-mode prefill --host ${local_ip} --port 30000 --trust-remote-code --dist-init-addr ${prefill_master_ip}:5000 --nnodes 2 --node-rank 1 --tp-size 16 --dp-size 8 --enable-dp-attention --enable-deepep-moe --deepep-mode normal --mem-fraction-static 0.8
# decode 0
$ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 ---disaggregation-transfer-backend nixl --disaggregation-mode decode --host ${local_ip} --port 30001 --trust-remote-code --dist-init-addr ${decode_master_ip}:5000 --nnodes 2 --node-rank 0 --tp-size 16 --dp-size 8 --enable-dp-attention --enable-deepep-moe --deepep-mode low_latency --mem-fraction-static 0.8 --max-running-requests 128
# decode 1
$ python -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 ---disaggregation-transfer-backend nixl --disaggregation-mode decode --host ${local_ip} --port 30001 --trust-remote-code --dist-init-addr ${decode_master_ip}:5000 --nnodes 2 --node-rank 1 --tp-size 16 --dp-size 8 --enable-dp-attention --enable-deepep-moe --deepep-mode low_latency --mem-fraction-static 0.8 --max-running-requests 128
```