[main][prefill optimization] Optimize parallel strategies to reduce communication overhead #2198

kunpengW-code · 2025-08-04T09:32:27Z

What this PR does / why we need it?

1.Shared Expert Sharding Strategy Update: Switched from TP-aligned to pure DP for shared experts, enabling more efficient execution.
2.O_Proj AllReduce → ReduceScatter: Reduced communication overhead by using ReduceScatter, made possible by pure DP sharding.
3.AllGather Postponed: Delayed to after QKV down projection to reduce synchronization impact during prefill.

Does this PR introduce any user-facing change?

How was this patch tested?

Adding ut case in tests/ut/attention/test_mla_v1.py

How to run

use parameter --additional_config='{"enable_shared_expert_dp": true}'

a.How to run eager mode

eg:
python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --enforce-eager --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true}'

b.How to run graph mode

eg:
python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true,"torchair_graph_config":{"enabled":true}}'

vLLM version: v0.10.0
vLLM main: vllm-project/vllm@9edd1db

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

github-actions · 2025-08-04T09:34:41Z

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

A PR should do only one thing, smaller PRs enable faster reviews.
Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

momo609 · 2025-08-04T10:13:10Z

vllm_ascend/models/deepseek_v2.py

-            kv_c, k_pe = self.kv_a_proj_with_mqa(hidden_states)[0].split(
+            kv_no_split = self.kv_a_proj_with_mqa(hidden_states)[0]
+            if self.enable_shared_expert_dp and self.debug_layer_idx > 3 and self.debug_layer_idx < 61:
+                hidden_states_or_q_c = get_tp_group().all_gather(


In the case of severe load imbalance between DPs, the DP domain may become stuck？

Here the DP domain will not block, this is all_gather of all TPs within the DP domain, and different DP domains will not affect each other.

…to upstream_main_shared_expert_dp

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

codecov · 2025-08-04T14:32:32Z

Codecov Report

❌ Patch coverage is 53.00000% with 47 lines in your changes missing coverage. Please review.
✅ Project coverage is 76.82%. Comparing base (875a86c) to head (0cf9e1d).
⚠️ Report is 28 commits behind head on main.

Files with missing lines	Patch %	Lines
vllm_ascend/models/deepseek_v2.py	27.90%	31 Missing ⚠️
vllm_ascend/attention/mla_v1.py	52.17%	11 Missing ⚠️
vllm_ascend/ops/fused_moe.py	55.55%	4 Missing ⚠️
vllm_ascend/ascend_config.py	66.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2198      +/-   ##
==========================================
+ Coverage   76.70%   76.82%   +0.11%     
==========================================
  Files         113      113              
  Lines       12944    13051     +107     
==========================================
+ Hits         9929    10026      +97     
- Misses       3015     3025      +10

Flag	Coverage Δ
unittests	`76.82% <53.00%> (+0.11%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

kunpengW-code · 2025-08-05T01:42:51Z

docs/source/user_guide/configuration/additional_config.md

 | `expert_map_path`             | str  | `None` | When using expert load balancing for the MOE model, an expert map path needs to be passed in. |
 | `chunked_prefill_for_mla`     | bool | `False` | Whether to enable the fused operator-like chunked_prefill. |
 | `kv_cache_dtype`     | str | `None` | When using the kv cache quantization method, kv cache dtype needs to be set, currently only int8 is supported. |
+| `enable_shared_expert_dp`     | bool | `True` | When the shared expert in DP, it has better performance but consumes more memory. When the memory is sensitive, this switch can be turned off manually. |


add the note，please take a look, thanks. @wangxiyuan

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

jianzs · 2025-08-05T11:25:08Z

Refine your title and description to better reflect your PR.

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

Signed-off-by: SlightwindSec <slightwindsec@gmail.com>

SlightwindSec · 2025-08-06T08:46:56Z

vllm_ascend/ops/fused_moe.py

        # to avoid accumulating too much tokens on a single rank.
        # currently it is only activated when doing profile runs.
-        if enable_force_load_balance and not self.use_aclgraph:
+        if enable_force_load_balance:


This line change appears to be unrelated to the purpose of this PR.

Signed-off-by: SlightwindSec <slightwindsec@gmail.com>

…d_expert_dp

wangxiyuan · 2025-08-07T11:46:22Z

@jianzs please take a look again.

jianzs · 2025-08-11T03:57:17Z

Are there any prerequisite features needed for this to function correctly? If yes, I recommend adding configuration validation to catch any missing dependencies early.

SlightwindSec · 2025-08-11T06:30:06Z

Are there any prerequisite features needed for this to function correctly? If yes, I recommend adding configuration validation to catch any missing dependencies early.

Thanks for your suggestion! This feature requires Torchair to be disabled and Expert Parallelism (EP) to be enabled.

self.enable_shared_expert_dp = additional_config.get(
            "enable_shared_expert_dp", True
        ) and not self.torchair_graph_config.enabled and vllm_config.parallel_config.enable_expert_parallel

I've already implemented a check for this in the code, so enable_shared_expert_dp will only be activated if these conditions are met. I've also added a warning that notifies users and disables the feature automatically if they try to use it in an unsupported configuration. @jianzs

* enable mm allreduce test (vllm-project#2192) ### What this PR does / why we need it? This PR is to add e2e test for using npu_mm_all_reduce_base fusion kernel. ### Does this PR introduce _any_ user-facing change? no ### How was this patch tested? not involved - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@5d5d419 Signed-off-by: Ronald1995 <ronaldautomobile@163.com> * [main] remove torch.cat and replace it by List[0] (vllm-project#2153) ### What this PR does / why we need it? torch_npu.npu_grouped_matmul: https://www.hiascend.com/document/detail/zh/Pytorch/710/apiref/torchnpuCustomsapi/context/torch_npu-npu_grouped_matmul.md According to the document, when `split_item` is 2 or 3, `torch_npu.npu_grouped_matmul` will return a list which has one element. Therefore, the `torch.cat` after `torch_npu.npu_grouped_matmul` is unnecessary. ### Does this PR introduce _any_ user-facing change? not involved ### How was this patch tested? ut and e2e covered: `tests/ut/ops/test_fused_ops.py`, `tests/e2e/singlecard/ops/test_fused_moe.py` **performance**: (qwen3 30B, 2k->20k) base: Total Token throughput (tok/s): 667.76 remove cat: Total Token throughput (tok/s): 680.82 - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@fa00c5d Signed-off-by: huangxialu <huangxialu1@huawei.com> * [CI][Quickfix] Fix AscendFusedMoE init error (vllm-project#2268) ### What this PR does / why we need it? Fix AscendFusedMoE init error. Use `super().__init__()` instead of `super(FusedMoE, self).__init__()` to ensure the member variables in base class could be called by the children class ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new existing test. - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@766bc81 --------- Signed-off-by: MengqingCao <cmq0113@163.com> * Fix accuracy test config and add DeepSeek-V2-Lite test (vllm-project#2261) ### What this PR does / why we need it? This PR fix accuracy test related to vllm-project#2073, users can now perform accuracy tests on multiple models simultaneously and generate different report files by running: ```bash cd ~/vllm-ascend pytest -sv ./tests/e2e/models/test_lm_eval_correctness.py \ --config-list-file ./tests/e2e/models/configs/accuracy.txt ``` ### Does this PR introduce _any_ user-facing change? no ### How was this patch tested? <img width="1648" height="511" alt="image" src="https://github.com/user-attachments/assets/1757e3b8-a6b7-44e5-b701-80940dc756cd" /> - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@766bc81 --------- Signed-off-by: Icey <1790571317@qq.com> * Fix accuracy test create PR (vllm-project#2274) ### What this PR does / why we need it? Fix create PR of accuracy test ### Does this PR introduce _any_ user-facing change? no ### How was this patch tested? Local testing: nv-action/vllm-benchmarks#87 - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@099c046 --------- Signed-off-by: Icey <1790571317@qq.com> * Add ut for test_communicator.py (vllm-project#2293) ### What this PR does / why we need it? Add ut for test_communicator.py - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@e5ebeeb Signed-off-by: yangqinghao-cmss <yangqinghao_yewu@cmss.chinamobile.com> * [CI] Fix broken CI (vllm-project#2302) 1. disable test_eagle_ccorrectness test, we'll reopen it once oom error fixed. 2. drop transformers version limit for main, since vLLM rely on >=4.55.0, see: vllm-project/vllm@65552b4 3. fix kv_connector_output bug, see: vllm-project/vllm@796bae0 - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@d1af8b7 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> * [2/N][Refactor] torchair model runner refactor (vllm-project#2204) There is lot of torchair code in model runner leading the code hard for maintenance. We'll create new torchair_model_runner to split torchair related logic. Following the workflow vllm-project#2203 What's this PR do: move `torchair` related logic into `_get_forward_metadata_across_dp` and override it in torchair model runner - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@1b99028 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> * [core] Support capture custom ops into aclgraph (vllm-project#2113) ### What this PR does / why we need it? Thanks to the PR vllm-project#426 make vllm-ascend support the aclgraph inference to reduce the host overhead. However, the capability of aclgraph strongly relies on the functionality provided by `torch.compile`, which is the key feature supported in torch 2.x . Therefore, capture custom op into aclgraph is only possible when it can be recognize and captured by `torch.compile`. In this PR, we register the meta implementation of current custom ops to enable the fx graph capture. And by doing that, insert those custom ops into aclgraph become a natural thing to the ascend runtime. ### Does this PR introduce _any_ user-facing change? No user face change. ### How was this patch tested? Tested in unittest, we will integrate the `rotary_embedding` op into a small custom model and use `torch.compile` and aclgraph to capture and replay it to verify its functionality. - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@1b99028 --------- Signed-off-by: ganyi <pleaplusone.gy@gmail.com> * Bump actions/download-artifact from 4 to 5 (vllm-project#2311) Bumps [actions/download-artifact](https://github.com/actions/download-artifact) from 4 to 5. - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@ebf7605 Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * [Perf][MTP] Optimize reject sampler in greedy situation. (vllm-project#2137) This PR port optimization in PR vllm-project#2002 to main and makes it cleaner. - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@afa5b7c --------- Signed-off-by: whx-sjtu <2952154980@qq.com> * [3/N][Refactor] torchair model runner refactor (vllm-project#2207) There is lot of torchair code in model runner leading the code hard for maintenance. We'll create new torchair_model_runner to split torchair related logic. Following the workflow vllm-project#2203, this is the first PR. What's this PR do: create common function `_build_attention_metadata` and `_generate_dummy_run_hidden_states` for dummy_run - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@ebf7605 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> * [Feat] chunkprefill mla support torchair graph (vllm-project#1772) chunkprefill mla only support eager mode now，we want to optimaze it by support torchair graph, the idea is simple, when all the request is running in decode, use torchair graph to deal with it, else when chunkprefill or prefill only, use the eager mode - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@ebf7605 Signed-off-by: haojiangzheng <justineric096@gmail.com> Co-authored-by: haojiangzheng <justineric096@gmail.com> * [4/N][Refactor] torchair model runner refactor (vllm-project#2208) There is lot of torchair code in model runner leading the code hard for maintenance. We'll create new torchair_model_runner to split torchair related logic. Following the workflow vllm-project#2203, this is the first PR. What's this PR do: create common function `_convert_torch_foramt` for initialize_kv_cache - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@14a5d90 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> * Configure Gemini (vllm-project#2298) ### What this PR does / why we need it? This PR requests Gemini AI to review PRs. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? NA - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@14a5d90 Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com> * ut: add ci guard for ut coverage (vllm-project#2317) ### What this PR does / why we need it? add ci guard for ut coverage, if ut coverage of patch pr is below 80%, the ci will failed/ ### Does this PR introduce _any_ user-facing change? not involved ### How was this patch tested? not involved - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@458e74e --------- Signed-off-by: Ronald1995 <ronaldautomobile@163.com> * [main][prefill optimization] Optimize parallel strategies to reduce communication overhead (vllm-project#2198) ### What this PR does / why we need it? 1.Shared Expert Sharding Strategy Update: Switched from TP-aligned to pure DP for shared experts, enabling more efficient execution. 2.O_Proj AllReduce → ReduceScatter: Reduced communication overhead by using ReduceScatter, made possible by pure DP sharding. 3.AllGather Postponed: Delayed to after QKV down projection to reduce synchronization impact during prefill. ### How was this patch tested? Adding ut case in `tests/ut/attention/test_mla_v1.py` #### How to run use parameter `--additional_config='{"enable_shared_expert_dp": true}'` ##### a.How to run eager mode eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --enforce-eager --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true}' ##### b.How to run graph mode eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true,"torchair_graph_config":{"enabled":true}}' - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@9edd1db --------- Signed-off-by: Wang Kunpeng <1289706727@qq.com> Signed-off-by: SlightwindSec <slightwindsec@gmail.com> Co-authored-by: SlightwindSec <slightwindsec@gmail.com> * [Doc] Update faq (vllm-project#2334) ### What this PR does / why we need it? - update determinitic calculation - update support device ### Does this PR introduce _any_ user-facing change? - Users should update ray and protobuf when using ray as distributed backend - Users should change to use `export HCCL_DETERMINISTIC=true` when enabling determinitic calculation ### How was this patch tested? N/A - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@ea1292a Signed-off-by: MengqingCao <cmq0113@163.com> * [5/N][Refactor] torchair model runner refactor (vllm-project#2216) There is lot of torchair code in model runner leading the code hard for maintenance. We'll create new torchair_model_runner to split torchair related logic. Following the workflow vllm-project#2203 What's this PR do: create common function `_capture_model` for capture_model - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@1891a26 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> * [1/N][Feat] Support MoE models with ACL Graph and refactor MoE communication logic (vllm-project#2125) ### What this PR does / why we need it? This PR refactors the MoE (Mixture of Experts) communication logic by introducing a strategy pattern. It defines an abstract base class, `MoECommMethod`, which encapsulates different communication strategies for MoE layers. By decoupling the MoE implementation from any single communication method, this change makes it simpler to add, replace, or optimize communication strategies in the future. Plan / Roadmap 1. Introduce `MoECommMethod`, implement `AllGatherImpl`, and adapt ACL Graph handling to cover all scenarios (this PR). 2. Implement `MC2CommImpl` and `AllToAllCommImpl` to optimize performance in specific scenarios. 3. Enable W8A8 / Int8 models to use `unified_fused_experts`. Other notes * Data-parallel (DP) communication currently does not work with vLLM's dispatch/combine mechanisms; an alternative approach is required to resolve this incompatibility. - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@f7ad6a1 --------- Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com> * [Doc] Add container image save/load FAQ for offline environments (vllm-project#2347) ### What this PR does / why we need it? Add Docker export/import guide for air-gapped environments ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? NA - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@d16aa3d Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com> * [Bugfix] fix the oom when chunkprefill with long context like 64k (vllm-project#2319) The attn mask was declared in the mla.py，we don't need the splitfuse mask when mla chunkprefill, and this mask will cause memory problem when long context like 64k or 128k - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@14a5d90 --------- Signed-off-by: haojiangzheng <justineric096@gmail.com> --------- Signed-off-by: Ronald1995 <ronaldautomobile@163.com> Signed-off-by: huangxialu <huangxialu1@huawei.com> Signed-off-by: MengqingCao <cmq0113@163.com> Signed-off-by: Icey <1790571317@qq.com> Signed-off-by: yangqinghao-cmss <yangqinghao_yewu@cmss.chinamobile.com> Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com> Signed-off-by: ganyi <pleaplusone.gy@gmail.com> Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: whx-sjtu <2952154980@qq.com> Signed-off-by: haojiangzheng <justineric096@gmail.com> Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com> Signed-off-by: Wang Kunpeng <1289706727@qq.com> Signed-off-by: SlightwindSec <slightwindsec@gmail.com> Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com> Co-authored-by: Ronald1995 <ronaldautomobile@163.com> Co-authored-by: huangxialu <huangxialu1@huawei.com> Co-authored-by: Mengqing Cao <cmq0113@163.com> Co-authored-by: Icey <1790571317@qq.com> Co-authored-by: yangqinghao-cmss <yangqinghao_yewu@cmss.chinamobile.com> Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com> Co-authored-by: Pleaplusone <pleaplusone.gy@gmail.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: whx <56632993+whx-sjtu@users.noreply.github.com> Co-authored-by: zhenghaojiang <zhjoneson@163.com> Co-authored-by: haojiangzheng <justineric096@gmail.com> Co-authored-by: jack <QwertyJack@users.noreply.github.com> Co-authored-by: Wang Kunpeng <1289706727@qq.com> Co-authored-by: SlightwindSec <slightwindsec@gmail.com> Co-authored-by: yiz-liu <136800916+yiz-liu@users.noreply.github.com>

…ommunication overhead (vllm-project#2198) 1.Shared Expert Sharding Strategy Update: Switched from TP-aligned to pure DP for shared experts, enabling more efficient execution. 2.O_Proj AllReduce → ReduceScatter: Reduced communication overhead by using ReduceScatter, made possible by pure DP sharding. 3.AllGather Postponed: Delayed to after QKV down projection to reduce synchronization impact during prefill. Adding ut case in `tests/ut/attention/test_mla_v1.py` use parameter `--additional_config='{"enable_shared_expert_dp": true}'` eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --enforce-eager --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true}' eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true,"torchair_graph_config":{"enabled":true}}' - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@9edd1db --------- Signed-off-by: Wang Kunpeng <1289706727@qq.com> Signed-off-by: SlightwindSec <slightwindsec@gmail.com> Co-authored-by: SlightwindSec <slightwindsec@gmail.com>

…ommunication overhead (vllm-project#2198) ### What this PR does / why we need it? 1.Shared Expert Sharding Strategy Update: Switched from TP-aligned to pure DP for shared experts, enabling more efficient execution. 2.O_Proj AllReduce → ReduceScatter: Reduced communication overhead by using ReduceScatter, made possible by pure DP sharding. 3.AllGather Postponed: Delayed to after QKV down projection to reduce synchronization impact during prefill. ### How was this patch tested? Adding ut case in `tests/ut/attention/test_mla_v1.py` #### How to run use parameter `--additional_config='{"enable_shared_expert_dp": true}'` ##### a.How to run eager mode eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --enforce-eager --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true}' ##### b.How to run graph mode eg: python -m vllm.entrypoints.openai.api_server --model=/model_path --trust-remote-code -tp 8 -dp 2 --enable_expert_parallel --port 8002 --max-model-len 5120 --max-num-batched-tokens 16384 --disable-log-requests --additional_config='{"ascend_scheduler_config":{"enabled":true},"enable_shared_expert_dp": true,"chunked_prefill_for_mla":true,"torchair_graph_config":{"enabled":true}}' - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@9edd1db --------- Signed-off-by: Wang Kunpeng <1289706727@qq.com> Signed-off-by: SlightwindSec <slightwindsec@gmail.com> Co-authored-by: SlightwindSec <slightwindsec@gmail.com>

[main][Prefill Perf] Parallel Strategy Optimizations

4ecbd33

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

github-actions bot added module:tests module:ops module:core labels Aug 4, 2025

momo609 reviewed Aug 4, 2025

View reviewed changes

kunpengW-code mentioned this pull request Aug 4, 2025

[WIP][main][Prefill Perf] Parallel Strategy Optimizations #2110

Closed

kunpengW-code added 2 commits August 4, 2025 22:06

Merge branch 'main' of https://github.com/vllm-project/vllm-ascend in…

63967cb

…to upstream_main_shared_expert_dp

[main][Prefill Perf] add doc

b165945

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

github-actions bot added the documentation Improvements or additions to documentation label Aug 4, 2025

kunpengW-code commented Aug 5, 2025

View reviewed changes

[main][Prefill Perf] simplify judgment conditions

5e9bb72

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

kunpengW-code added 2 commits August 5, 2025 21:51

[main][Prefill Perf] fix ci

31bd0d7

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

[main][Prefill Perf] fix ci

19ac51c

Signed-off-by: Wang Kunpeng <1289706727@qq.com>

kunpengW-code changed the title ~~[main][Prefill Perf] Parallel Strategy Optimizations~~ [main][prefill optimization] Optimize parallel strategies to reduce communication overhead Aug 5, 2025

fix hardcode issue

b00fa60

Signed-off-by: SlightwindSec <slightwindsec@gmail.com>

SlightwindSec reviewed Aug 6, 2025

View reviewed changes

fix ut error

a4117fb

Signed-off-by: SlightwindSec <slightwindsec@gmail.com>

SlightwindSec force-pushed the upstream_main_shared_expert_dp branch from bd22a0d to a4117fb Compare August 6, 2025 11:39

Merge remote-tracking branch 'upstream/main' into upstream_main_share…

0cf9e1d

…d_expert_dp

wangxiyuan approved these changes Aug 7, 2025

View reviewed changes

wangxiyuan merged commit dc585f1 into vllm-project:main Aug 12, 2025
25 checks passed

SlightwindSec mentioned this pull request Sep 18, 2025

[RFC]: Prefill Performance Optimization for DeepSeek Large Scale EP #3012

Open

SlightwindSec deleted the upstream_main_shared_expert_dp branch October 13, 2025 01:42

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[main][prefill optimization] Optimize parallel strategies to reduce communication overhead #2198

[main][prefill optimization] Optimize parallel strategies to reduce communication overhead #2198

Uh oh!

kunpengW-code commented Aug 4, 2025 •

edited by github-actions bot

Loading

Uh oh!

github-actions bot commented Aug 4, 2025

Uh oh!

momo609 Aug 4, 2025

Uh oh!

kunpengW-code Aug 4, 2025

Uh oh!

codecov bot commented Aug 4, 2025 •

edited

Loading

Uh oh!

kunpengW-code Aug 5, 2025

Uh oh!

jianzs commented Aug 5, 2025 •

edited

Loading

Uh oh!

SlightwindSec Aug 6, 2025

Uh oh!

wangxiyuan commented Aug 7, 2025

Uh oh!

jianzs commented Aug 11, 2025

Uh oh!

SlightwindSec commented Aug 11, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

[main][prefill optimization] Optimize parallel strategies to reduce communication overhead #2198

[main][prefill optimization] Optimize parallel strategies to reduce communication overhead #2198

Uh oh!

Conversation

kunpengW-code commented Aug 4, 2025 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What this PR does / why we need it?

Does this PR introduce any user-facing change?

How was this patch tested?

How to run

a.How to run eager mode

b.How to run graph mode

Uh oh!

github-actions bot commented Aug 4, 2025

Uh oh!

momo609 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

kunpengW-code Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Aug 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

kunpengW-code Aug 5, 2025

Choose a reason for hiding this comment

Uh oh!

jianzs commented Aug 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

SlightwindSec Aug 6, 2025

Choose a reason for hiding this comment

Uh oh!

wangxiyuan commented Aug 7, 2025

Uh oh!

jianzs commented Aug 11, 2025

Uh oh!

SlightwindSec commented Aug 11, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

kunpengW-code commented Aug 4, 2025 •

edited by github-actions bot

Loading

codecov bot commented Aug 4, 2025 •

edited

Loading

jianzs commented Aug 5, 2025 •

edited

Loading