50 Commits
Author SHA1 Message Date
Rui Qiao 09a6a49eaf [Misc] Avoid "PyTorch non-writable tensors" warning in RayPPCommunicator (#27443)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-10-24 14:53:09 +08:00
Rui Qiao 757fa4a4da [DP][ray] Support different VLLM_RAY_DP_PACK_STRATEGY (#23849)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-10-09 19:53:43 -07:00
Rui Qiao 467a4f98f1 [Misc] Redact ray runtime env before logging (#26302)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-10-08 17:43:34 -07:00
Rui Qiao 302962e806 [Bugfix] Skip dead and non-GPU nodes for Ray DP engine allocation (#22275)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-08-05 20:35:32 -07:00
Rui Qiao 2ff46b8826 [Misc] Bump ray to 2.48.0 (#22123)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-08-02 19:42:00 -07:00
Rui Qiao 4ac8437352 [Misc] Getting and passing ray runtime_env to workers (#22040)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-08-01 23:54:40 -07:00
Rui Qiao d331759488 Introduce RayPPCommunicator for ray-based PP (#21660)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-08-01 11:50:58 -07:00
Rui Qiao c7742d6113 [Bugfix] Always set RAY_ADDRESS for Ray actor before spawn (#21540)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-25 17:08:30 -07:00
Rui Qiao cea96a0156 [Bugfix] Fix sync_and_slice_intermediate_tensors (#21537)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-25 17:07:58 -07:00
Rui Qiao 1e9ea8e69d [P/D] Move FakeNixlWrapper to test dir (#21328)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-24 08:53:45 -07:00
Rui Qiao 73e3949d07 [Misc] Improve comment for DPEngineCoreActor._set_cuda_visible_devices() (#21501)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-24 03:13:40 -07:00
Rui Qiao 35bc8bd5fb [Misc] Copy HF_TOKEN env var to Ray workers (#21406)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-22 16:18:42 -07:00
Rui Qiao 217937221b Elastic Expert Parallel Initial Support (#20775)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-18 17:46:09 -07:00
Rui Qiao d91278181d [doc] Add more details for Ray-based DP (#20948)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-15 05:37:12 -07:00
Rui Qiao a6d795d593 [DP] Copy environment variables to Ray DPEngineCoreActors (#20344)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-07-07 10:14:22 -07:00
Rui Qiao bdce64f236 [V1] Support DP with Ray (#18779) 2025-06-02 21:15:13 -07:00
Rui Qiao c44c384b1c [Misc] Add references in ray_serve_deepseek example (#17907)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-05-09 16:59:36 +00:00
Rui Qiao c53e0730cb [Misc] Refine ray_serve_deepseek example (#17204)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-04-25 16:06:59 -07:00
Rui Qiao 583e900996 [Misc] Add example to run DeepSeek with Ray Serve LLM (#17134)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-04-24 22:25:21 +00:00
Rui Qiao c0dfd97519 [V1][PP] Optimization: continue scheduling prefill chunks (#17080)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-04-24 05:27:08 -07:00
8dd41d6bcc [Misc] Use envs.VLLM_USE_RAY_COMPILED_DAG_CHANNEL_TYPE (#15831)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
2025-04-01 06:07:53 -07:00
Rui Qiao df8d3d1287 [Misc] Restrict ray version dependency and update PP feature warning in V1 (#15556) 2025-03-27 06:21:07 +00:00
Rui Qiao 5797fb97e9 [Misc] Remove ignore_reinit_error for ray.init() (#15373) 2025-03-24 07:41:53 +00:00
Rui Qiao 4c69e228b3 [Misc] Increase RayDistributedExecutor RAY_CGRAPH_get_timeout (#15301)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-03-21 22:25:43 -07:00
Rui Qiao 4cb1c05c9e [Doc] Clarify run vllm only on one node in distributed inference (#15148)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-03-20 09:55:59 +08:00
Rui Qiao b9b5bdfc7d [Misc] Catching Ray Compiled Graph PP test failures for V1 (#14847) 2025-03-16 15:46:42 -07:00
Rui Qiao abcc61e0af [misc] Mention ray list nodes command to troubleshoot ray issues (#14318)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-03-06 02:00:36 +00:00
Rui Qiao bf13d40972 [core] Pass all driver env vars to ray workers unless excluded (#14099)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-03-04 11:44:17 +08:00
Rui Qiao 084bbac8cc [core] Bump ray to 2.43 (#13994)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-02-28 21:47:44 +00:00
Rui Qiao 06c8f8d885 [bugfix] Fix profiling for RayDistributedExecutor (#13945)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-02-28 01:01:21 +08:00
Rui Qiao c9944acbf9 [misc] Rename Ray ADAG to Compiled Graph (#13928) 2025-02-26 20:03:28 -08:00
Rui Qiao 9ba28043b5 [misc] Show driver IP info when Ray fails to allocate driver worker (#13858)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-02-26 09:53:43 +08:00
Rui Qiao 3173c3b34e [misc] Clean up ray compiled graph type hints (#13731) 2025-02-25 00:37:08 -08:00
Rui Qiao 9605c1256e [V1][core] Implement pipeline parallel on Ray (#12996) 2025-02-13 08:02:46 +00:00
Rui Qiao f218f9c24d [core] Turn off GPU communication overlap for Ray executor (#12051)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-01-15 05:19:55 +00:00
Rui Qiao 022c5c6944 [V1] Refactor get_executor_cls (#11754) 2025-01-06 07:59:16 +00:00
Rui Qiao f8fcca100b [Misc] Fix typo for valid_tool_parses (#11753)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2025-01-06 07:12:38 +00:00
Rui Qiao 9832e5572a [V1] Unify VLLM_ENABLE_V1_MULTIPROCESSING handling in RayExecutor (#11472) 2024-12-24 19:49:46 -08:00
Rui Qiao a491d6f535 [V1] TP Ray executor (#11107)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-12-23 23:00:12 +00:00
Rui QiaoandCody Yu f26c4aeecb [Misc] Optimize ray worker initialization time (#11275)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>
2024-12-18 23:38:02 -08:00
Rui QiaoandRui Qiao 72ff3a9686 [core] Bump ray to use _overlap_gpu_communication in compiled graph tests (#10410)
Signed-off-by: Rui Qiao <ubuntu@ip-172-31-15-128.us-west-2.compute.internal>
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
Co-authored-by: Rui Qiao <ubuntu@ip-172-31-15-128.us-west-2.compute.internal>
2024-12-11 11:36:35 -08:00
Rui Qiao cbdb252259 [Misc] Limit to ray[adag] 2.35 to avoid backward incompatible change (#8509)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-09-17 00:06:26 -07:00
Rui Qiao de80783b69 [Misc] Use ray[adag] dependency instead of cuda (#7938) 2024-09-06 09:18:35 -07:00
Rui Qiao bae888cb8e [Bugfix] Clear engine reference in AsyncEngineRPCServer (#7618)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-16 20:44:05 -07:00
Rui Qiao 198d6a2898 [Core] Shut down aDAG workers with clean async llm engine exit (#7224)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-12 17:57:16 -07:00
Rui Qiao 746709642c [Misc] Fix typos in scheduler.py (#7285)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-07 17:06:01 -07:00
Rui Qiao 997cf78308 [Misc] Fix typo in GroupCoordinator.recv() (#7167)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-05 11:10:16 -07:00
Rui Qiao 22e718ff1a [Misc] Revive to use loopback address for driver IP (#7091)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-02 15:50:00 -07:00
Rui Qiao 05308891e2 [Core] Pipeline parallel with Ray ADAG (#6837)
Support pipeline-parallelism with Ray accelerated DAG.

Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
2024-08-02 13:55:40 -07:00
Rui QiaoandStephanie Wang 61e592747c [Core] Introduce SPMD worker execution using Ray accelerated DAG (#6032)
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
Co-authored-by: Stephanie Wang <swang@cs.berkeley.edu>
2024-07-17 22:27:09 -07:00