EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Doc]clean up ascend scheduler config from doc (#4612)

Browse Source 
clean up ascend scheduler config from doc

- vLLM version: v0.11.2

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-12-02 14:22:56 +08:00
committed by GitHub
parent bb1610dc25
commit cb33b09179
8 changed files with 13 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
docs/source/tutorials/DeepSeek-V3.2-Exp.md
View File

@@ -174,7 +174,7 @@ vllm serve vllm-ascend/DeepSeek-V3.2-Exp-W8A8 \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.92 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
--additional-config '{"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
```
### Multi-node Deployment
@@ -226,7 +226,7 @@ vllm serve /root/.cache/Modelers_Park/DeepSeek-V3.2-Exp \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.9 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
--additional-config '{"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
```
**Node 1**
@@ -270,7 +270,7 @@ vllm serve /root/.cache/Modelers_Park/DeepSeek-V3.2-Exp \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.92 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
--additional-config '{"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
```
::::
@@ -318,7 +318,7 @@ vllm serve vllm-ascend/DeepSeek-V3.2-Exp-W8A8 \
--quantization ascend \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.9 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
--additional-config '{"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
```
**Node 1**
@@ -365,7 +365,7 @@ vllm serve vllm-ascend/DeepSeek-V3.2-Exp-W8A8 \
--quantization ascend \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.92 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
--additional-config '{"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}'
```
::::

4
docs/source/tutorials/multi_node.md
View File

@@ -137,7 +137,7 @@ vllm serve vllm-ascend/DeepSeek-V3.1-W8A8 \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.9 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true}}'
--additional-config '{"torchair_graph_config":{"enabled":true}}'
```
**Node 1**
@@ -182,7 +182,7 @@ vllm serve vllm-ascend/DeepSeek-V3.1-W8A8 \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.92 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true}}'
--additional-config '{"torchair_graph_config":{"enabled":true}}'
```
The deployment view looks like:

4
docs/source/tutorials/multi_node_kimi.md
View File

@@ -93,7 +93,7 @@ vllm serve /home/cache/weights/Kimi-K2-Instruct-W8A8 \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.9 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true}}'
--additional-config '{"torchair_graph_config":{"enabled":true}}'
```
**Node 1**
@@ -137,7 +137,7 @@ vllm serve /home/cache/weights/Kimi-K2-Instruct-W8A8 \
--trust-remote-code \
--no-enable-prefix-caching \
--gpu-memory-utilization 0.92 \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true}}'
--additional-config '{"torchair_graph_config":{"enabled":true}}'
```
The deployment view looks like:

7
docs/source/tutorials/multi_npu_moge.md
View File

@@ -157,12 +157,7 @@ if __name__ == "__main__":
additional_config={
'torchair_graph_config': {
'enabled': True,
},
'ascend_scheduler_config':{
'enabled': True,
'enable_chunked_prefill' : False,
'chunked_prefill_enabled': False
},
}
})
outputs = llm.generate(prompts, sampling_params)

19
docs/source/user_guide/configuration/additional_config.md
View File

@@ -27,7 +27,6 @@ The following table lists additional configuration options available in vLLM Asc
| Name | Type | Default | Description |
|-------------------------------------|------|---------|-----------------------------------------------------------------------------------------------------------------------------------------------|
| `torchair_graph_config` | dict | `{}` | Configuration options for torchair graph mode |
| `ascend_scheduler_config` | dict | `{}` | Configuration options for ascend scheduler |
| `weight_prefetch_config` | dict | `{}` | Configuration options for weight prefetch |
| `refresh` | bool | `false` | Whether to refresh global Ascend configuration content. This is usually used by rlhf or ut/e2e test case. |
| `expert_map_path` | str | `None` | When using expert load balancing for an MoE model, an expert map path needs to be passed in. |
@@ -61,18 +60,6 @@ The details of each configuration option are as follows:
| `enable_kv_nz`| bool | `False` | Whether to enable KV Cache NZ layout. This option only takes effect on models using MLA (for example, DeepSeek). |
| `enable_super_kernel` | bool | `False` | Whether to enable super kernel to fuse operators in deepseek moe layers. This option only takes effects on moe models using dynamic w8a8 quantization.|
**ascend_scheduler_config**
| Name | Type | Default | Description |
| ---- | ---- | ------- | ----------- |
| `enabled` | bool | `False` | Whether to enable ascend scheduler for V1 engine.|
| `enable_pd_transfer` | bool | `False` | Whether to enable P-D transfer. When it is enabled, decode is started only when prefill of all requests is done. This option only takes effect on offline inference. |
| `decode_max_num_seqs` | int | `0` | Whether to change max_num_seqs of decode phase when P-D transfer is enabled. This option only takes effect when enable_pd_transfer is True. |
| `max_long_partial_prefills` | Union[int, float] | `float('inf')` | The maximum number of prompts longer than long_prefill_token_threshold that will be prefilled concurrently. |
| `long_prefill_token_threshold` | Union[int, float] | `float('inf')` | a request is considered long if the prompt is longer than this number of tokens. |
ascend_scheduler_config also supports the options from [vllm scheduler config](https://docs.vllm.ai/en/stable/api/vllm/config.html#vllm.config.SchedulerConfig). For example, you can add `enable_chunked_prefill: True` to ascend_scheduler_config as well.
**weight_prefetch_config**
| Name | Type | Default | Description |
@@ -93,12 +80,6 @@ An example of additional configuration is as follows:
"graph_batch_sizes_init": False,
"enable_kv_nz": False
},
"ascend_scheduler_config": {
"enabled": True,
"enable_chunked_prefill": True,
"max_long_partial_prefills": 1,
"long_prefill_token_threshold": 4096,
},
"weight_prefetch_config": {
"enabled": True,
"prefetch_ratio": {

4
docs/source/user_guide/feature_guide/graph_mode.md
View File

@@ -45,14 +45,14 @@ import os
from vllm import LLM
# TorchAirGraph only works without chunked-prefill now
model = LLM(model="path/to/DeepSeek-R1-0528", additional_config={"torchair_graph_config": {"enabled": True},"ascend_scheduler_config": {"enabled": True}})
model = LLM(model="path/to/DeepSeek-R1-0528", additional_config={"torchair_graph_config": {"enabled": True}})
outputs = model.generate("Hello, how are you?")
```
Online example:
```shell
vllm serve path/to/DeepSeek-R1-0528 --additional-config='{"torchair_graph_config": {"enabled": true},"ascend_scheduler_config": {"enabled": true}}'
vllm serve path/to/DeepSeek-R1-0528 --additional-config='{"torchair_graph_config": {"enabled": true}}'
```
You can find more details about additional configuration [here](../configuration/additional_config.md).

1
examples/offline_inference_npu_long_seq.py
View File

@@ -42,7 +42,6 @@ if __name__ == "__main__":
enable_chunked_prefill=False,
max_num_batched_tokens=2048,
max_model_len=1024,
additional_config={"ascend_scheduler_config": {"enabled": False}},
max_num_seqs=1,
block_size=128,
gpu_memory_utilization=0.9

2
examples/run_dp_server.sh
View File

@@ -28,4 +28,4 @@ vllm serve Qwen/Qwen1.5-MoE-A2.7B \
--gpu-memory-utilization 0.9 \
--trust-remote-code \
--enforce-eager \
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":false, "use_cached_graph":false}}'
--additional-config '{"torchair_graph_config":{"enabled":false, "use_cached_graph":false}}'