biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Feat][Executor] Introduce RayExecutorV2 (#36836)

Browse Source 
Signed-off-by: Jeffrey Wang <jeffreywang@anyscale.com>
This commit is contained in:
Jeffrey Wang
2026-04-01 14:34:29 -07:00
committed by GitHub
parent cb268e4e55
commit de5e6c44c6
14 changed files with 1603 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
.buildkite/test_areas/distributed.yaml
View File

@@ -224,6 +224,20 @@ steps:
commands:
- ./.buildkite/scripts/run-multi-node-test.sh /vllm-workspace/tests 2 2 $IMAGE_TAG "VLLM_TEST_SAME_HOST=0 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_same_node.py | grep 'Same node test passed' && NUM_NODES=2 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_node_count.py | grep 'Node count test passed' && python3 ../examples/offline_inference/data_parallel.py -dp=2 -tp=1 --dp-num-nodes=2 --dp-node-rank=0 --dp-master-addr=192.168.10.10 --dp-master-port=12345 --enforce-eager --trust-remote-code && VLLM_MULTI_NODE=1 pytest -v -s distributed/test_multi_node_assignment.py && VLLM_MULTI_NODE=1 pytest -v -s distributed/test_pipeline_parallel.py" "VLLM_TEST_SAME_HOST=0 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_same_node.py | grep 'Same node test passed' && NUM_NODES=2 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_node_count.py | grep 'Node count test passed' && python3 ../examples/offline_inference/data_parallel.py -dp=2 -tp=1 --dp-num-nodes=2 --dp-node-rank=1 --dp-master-addr=192.168.10.10 --dp-master-port=12345 --enforce-eager --trust-remote-code"
- label: MessageQueue TCP Multi-Node (2 GPUs)
timeout_in_minutes: 10
working_dir: "/vllm-workspace/tests"
num_devices: 1
num_nodes: 2
no_plugin: true
optional: true
source_file_dependencies:
- vllm/distributed/device_communicators/shm_broadcast.py
- vllm/distributed/parallel_state.py
- tests/distributed/test_mq_tcp_multinode.py
commands:
- ./.buildkite/scripts/run-multi-node-test.sh /vllm-workspace/tests 2 1 $IMAGE_TAG "torchrun --nnodes 2 --nproc-per-node=1 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_mq_tcp_multinode.py" "torchrun --nnodes 2 --nproc-per-node=1 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_mq_tcp_multinode.py"
- label: Distributed NixlConnector PD accuracy (4 GPUs)
timeout_in_minutes: 30
working_dir: "/vllm-workspace/tests"
@@ -294,3 +308,23 @@ steps:
commands:
- pytest -v -s distributed/test_pp_cudagraph.py
- pytest -v -s distributed/test_pipeline_parallel.py
- label: RayExecutorV2 (4 GPUs)
timeout_in_minutes: 60
working_dir: "/vllm-workspace/tests"
num_devices: 4
source_file_dependencies:
- vllm/v1/executor/ray_executor_v2.py
- vllm/v1/executor/abstract.py
- vllm/v1/executor/multiproc_executor.py
- tests/distributed/test_ray_v2_executor.py
- tests/distributed/test_ray_v2_executor_e2e.py
- tests/distributed/test_pipeline_parallel.py
- tests/basic_correctness/test_basic_correctness.py
commands:
- export VLLM_USE_RAY_V2_EXECUTOR_BACKEND=1
- export NCCL_CUMEM_HOST_ENABLE=0
- pytest -v -s distributed/test_ray_v2_executor.py
- pytest -v -s distributed/test_ray_v2_executor_e2e.py
- pytest -v -s distributed/test_pipeline_parallel.py -k "ray"
- TARGET_TEST_SUITE=L4 pytest -v -s basic_correctness/test_basic_correctness.py -k "ray"