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

[CI] Add nightly test for Qwen3-235B-A22B with mooncake layerwise connector (#5441)

Browse Source 
### What this PR does / why we need it?

Add nightly test for Qwen3-235B-A22B with  mooncake layerwise connector.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
81786c8774

---------

Signed-off-by: wjunLu <wjunlu217@gmail.com>
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
Co-authored-by: hfadzxy <starmoon_zhang@163.com>
This commit is contained in:
wjunLu
2026-02-27 16:31:02 +08:00
committed by GitHub
parent c13d90b766
commit b60b991005
2 changed files with 118 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.github/workflows/schedule_nightly_test_a3.yaml vendored
View File

@@ -71,6 +71,9 @@ jobs:
- name: multi-node-dpsk3.2-2node
config_file_path: DeepSeek-V3_2-W8A8-A3-dual-nodes.yaml
size: 2
- name: multi-node-qwen3-dp-mooncake-layerwise
config_file_path: Qwen3-235B-A22B-Mooncake-Layerwise.yaml
size: 2
- name: multi-node-deepseek-r1-w8a8-longseq
config_file_path: DeepSeek-R1-W8A8-longseq.yaml
size: 2

115
tests/e2e/nightly/multi_node/config/Qwen3-235B-A22B-Mooncake-Layerwise.yaml Normal file
View File

@@ -0,0 +1,115 @@
test_name: "test Qwen3-235B-A22B PD separation with mooncake layerwise connector"
model: "Qwen/Qwen3-235B-A22B"
num_nodes: 2
npu_per_node: 16
env_common:
VLLM_USE_MODELSCOPE: true
OMP_PROC_BIND: false
OMP_NUM_THREADS: 100
HCCL_BUFFSIZE: 1024
SERVER_PORT: 8080
NUMEXPR_MAX_THREADS: 128
MC_LOG_LEVEL: "ERROR"
DISAGGREGATED_PREFILL_PROXY_SCRIPT: "examples/disaggregated_prefill_v1/load_balance_proxy_layerwise_server_example.py"
disaggregated_prefill:
enabled: true
prefiller_host_index: [0]
decoder_host_index: [1]
deployment:
-
server_cmd: >
vllm serve "Qwen/Qwen3-235B-A22B"
--host 0.0.0.0
--port $SERVER_PORT
--data-parallel-size 2
--data-parallel-size-local 2
--data-parallel-address $LOCAL_IP
--data-parallel-rpc-port 13389
--tensor-parallel-size 8
--seed 1024
--enforce-eager
--enable-expert-parallel
--max-num-seqs 32
--max-model-len 8192
--max-num-batched-tokens 8192
--trust-remote-code
--no-enable-prefix-caching
--gpu-memory-utilization 0.9
--additional-config '{"recompute_scheduler_enable": true,"enable_shared_expert_dp": true}'
--kv-transfer-config
'{"kv_connector": "MooncakeLayerwiseConnector",
"kv_role": "kv_producer",
"kv_port": "30000",
"engine_id": "0",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_layerwise_connector",
"kv_connector_extra_config": {
"prefill": {
"dp_size": 2,
"tp_size": 8
},
"decode": {
"dp_size": 2,
"tp_size": 8
}
}
}'
-
server_cmd: >
vllm serve "Qwen/Qwen3-235B-A22B"
--host 0.0.0.0
--port $SERVER_PORT
--data-parallel-size 2
--data-parallel-size-local 2
--data-parallel-address $MASTER_IP
--data-parallel-rpc-port 13389
--tensor-parallel-size 8
--seed 1024
--max-num-seqs 16
--max-model-len 8192
--max-num-batched-tokens 8192
--enable-expert-parallel
--trust-remote-code
--no-enable-prefix-caching
--gpu-memory-utilization 0.9
--additional-config '{"torchair_graph_config":{"enabled":true}}'
--kv-transfer-config
'{"kv_connector": "MooncakeLayerwiseConnector",
"kv_role": "kv_consumer",
"kv_port": "30200",
"engine_id": "1",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_layerwise_connector",
"kv_connector_extra_config": {
"prefill": {
"dp_size": 2,
"tp_size": 8
},
"decode": {
"dp_size": 2,
"tp_size": 8
}
}
}'
benchmarks:
perf:
case_type: performance
dataset_path: vllm-ascend/GSM8K-in3500-bs2800
request_conf: vllm_api_stream_chat
dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
num_prompts: 2800
max_out_len: 1500
batch_size: 700
request_rate: 11.2
baseline: 183
threshold: 0.97
acc:
case_type: accuracy
dataset_path: vllm-ascend/gsm8k-lite
request_conf: vllm_api_general_chat
dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_chat_prompt
max_out_len: 7680
batch_size: 512
baseline: 97
threshold: 3