Commit Graph

67 Commits

Author SHA1 Message Date
ybyang
fa8993111d Fix: Add token heuristic increment in total_tokens load balancing (#22614) 2026-04-21 01:29:33 -07:00
Liangsheng Yin
573e12a7fc Merge /get_load into /v1/loads (#23010) 2026-04-17 13:36:51 -07:00
Bi Xue
c43716a357 [sgl] provide an option to send control req to all dp ranks rank0 (#22758) 2026-04-16 14:24:26 +08:00
Yilong Zhao
fe92f3563c dp: add profile req hook (#22083) 2026-04-03 20:47:09 -07:00
Shiyan Deng
0c9dc098e7 Fix DP attention worker port binding for IPv6 support (#21917)
Signed-off-by: Shiyan Deng <dsy842974287@meta.com>
2026-04-03 12:39:39 -07:00
Yilong Zhao
95327458ee misc: add BatchTokenizerReq hook into dp controller (#20981) 2026-03-19 23:59:53 -07:00
psaab
9f56b471aa [Network] Use NetworkAddress for dist_init_method and loopback fallbacks (#20657)
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
2026-03-16 19:59:49 -07:00
Liangsheng Yin
f0458e0b49 [Utils] Move network/socket utilities from common.py to network.py (#20646) 2026-03-15 20:35:24 -07:00
Liangsheng Yin
61cad15d28 [Utils] Add NetworkAddress abstraction for IPv6-safe address handling (#20306) 2026-03-11 00:07:37 -07:00
Liangsheng Yin
539f772f54 [PD-Disagg] Fully support external DP dispatch w/ PD-disaggregation mode. (#19268)
Co-authored-by: Ratish P <114130421+ratish1@users.noreply.github.com>
2026-02-24 19:58:01 -08:00
Feng Su
3b89302277 Refactor: observability code cleanup (#17862)
Signed-off-by: Feng Su <sufeng@linux.alibaba.com>
2026-02-24 18:07:29 -08:00
Estrella-xx
1b3513a7e4 refactor FAKE transfer backend and remove --disaggregation-decode-enable-fake-auto parameter (#18345) 2026-02-16 17:27:02 +03:00
dongjiyingdjy
8b4c364960 refactor context parallel state (#17213)
Co-authored-by: Shunkang <182541032+Shunkangz@users.noreply.github.co>
2026-02-13 23:18:17 +08:00
ympcMark
f7a5e425c3 [3/N] Achieve fault tolerance at the DP level (#11657)
Co-authored-by: UNIDY <unidy2002@outlook.com>
Co-authored-by: Hank Han <hanhan7630@outlook.com>
2026-01-20 18:47:08 +08:00
Ratish P
c0248d6f37 [dpc]: unify DP controller load balancing and simplify dispatch logic (#16258) 2026-01-11 12:38:03 +08:00
fzyzcjy
32a569fb77 Tiny add CPU resource monitoring for overload diagnosis (#16852) 2026-01-10 12:48:18 +08:00
Yuhao Yang
2138ff48c6 Revert "[FEAT] optimize tensor zmq transfer for multimodal inputs" (#16386) 2026-01-04 22:05:26 -08:00
siyu
078d96213a [FEAT] optimize tensor zmq transfer for multimodal inputs (#13592)
Co-authored-by: Shangming Cai <csmthu@gmail.com>
Co-authored-by: Yuhao Yang <47235274+yhyang201@users.noreply.github.com>
2026-01-03 12:23:23 +08:00
Mufeez Amjad
cbff7ad985 dp-attention: add follow_bootstrap_room + auto load-balance; drop decode_round_robin (#16110) 2025-12-30 22:33:06 +08:00
Lianmin Zheng
e6d5a213ad Fix metrics (#15998) 2025-12-28 05:03:49 -08:00
Huaixin Chang
0c39730b18 DP: support piggyback server load report (#11469)
Signed-off-by: Chang Huaixin (OpenAnolis) <changhuaixin@linux.alibaba.com>
2025-12-25 11:35:05 +08:00
Baidu-AIAK
bc3ca30023 [PD] Support fake decode for PD disaggregation without prefill node (#14628)
Co-authored-by: sunhailiang <sunhailiang@baidu.com>
2025-12-23 12:43:33 +08:00
fzyzcjy
d5431ff894 Tiny add stuck simulation (#15613) 2025-12-22 17:00:18 +08:00
fzyzcjy
454a2544f2 Support soft watchdog for tokenizer/detokenizer/dp-controller processes (#15607) 2025-12-22 16:59:16 +08:00
Hexq0210
cb30d056e3 add decode round robin policy (#15164) 2025-12-22 14:48:52 +08:00
Lianmin Zheng
4285e99da7 [Auto Sync] Update data_parallel_controller.py, detokenizer... (20251209) (#14759)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-12-09 18:38:38 -08:00
fzyzcjy
340c613ab5 Support numactl bind for CPU and memory before process starts (#14156) 2025-11-30 17:00:33 +08:00
Lianmin Zheng
ae62279071 Fix data parallel controller launch for num nodes > 2 (#12822) 2025-11-07 12:13:21 -08:00
Feng Su
ea96106000 [Feature] Sglang Tracing: Fine-Grained Tracking for Request Latency - Part 2 (#10804)
Signed-off-by: Feng Su <sufeng@linux.alibaba.com>
2025-10-28 01:25:46 -07:00
fzyzcjy
ea6275dfbc Tiny add hints when users send requests to wrong place (#11808) 2025-10-19 16:10:20 +08:00
Lianmin Zheng
9eefe2c0b7 Set CUDA_VISIBLE_DEVICES to achieve one GPU per process (#9170)
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: Cheng Wan <cwan@x.ai>
Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>
2025-10-17 17:30:06 -07:00
Yongtong Wu
a20e7df8d0 Improve dp attention port assignment scheme (#5889)
Co-authored-by: Cheng Wan <cwan@x.ai>
2025-10-12 17:55:59 -07:00
Lianmin Zheng
9b8ebb2798 move more files under srt/utils (#11285) 2025-10-09 16:46:15 -07:00
XSongQ
546914fa2d [Fix] Fix the bug of the calculation of base_gpu_id (dp offset) in data_parallel_controller.py (#10741) 2025-10-03 13:25:57 -07:00
Liangsheng Yin
7ff740a6ce Remove dp balance metadata and minimul token balance. (#11170) 2025-10-03 01:48:15 +08:00
Liangsheng Yin
305c9e8c2d [4/N]DP refactor: support watching mode get_load and shortest queue strategy (#10201) 2025-09-15 10:06:08 +08:00
Liangsheng Yin
83d55ac51f [1/N]DP refactor: Improve dp rank scheduling in PD disaggregation mode. (#10169) 2025-09-09 12:27:55 +08:00
Caproni
96784a65fd [Fix] Orphan process in data parallel (#7995)
Signed-off-by: Capronir <839972205@qq.com>
2025-09-09 11:09:09 +08:00
Guanhua Wang
f7b2853ff8 [feat] support minimum token load balance in dp attention (#7379) 2025-08-03 00:46:47 -07:00
Cheng Wan
7a1f7fc504 [Feature] Hybrid EP and TP (#8590) 2025-07-31 02:53:25 -07:00
fzyzcjy
0ce84c822b Support colocating requests (#7973) 2025-07-28 22:51:49 -07:00
ishandhanani
f1569876d5 feat: add direct routing strategy to DP worker (#6884) 2025-06-09 11:44:05 -07:00
applesaucethebun
2ce8793519 Add typo checker in pre-commit (#6179)
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
2025-05-11 12:55:00 +08:00
Lianmin Zheng
4319978c73 Fix data parallel perf regression (#6183) 2025-05-10 19:18:35 -07:00
Ying Sheng
11383cec3c [PP] Add pipeline parallelism (#5724) 2025-04-30 18:18:07 -07:00
Cheng Wan
711efe7814 Integrating PD disaggregation with DP attention and DeepEP (#5435)
Co-authored-by: Byron Hsu <byronhsu1230@gmail.com>
2025-04-23 01:46:01 -07:00
fzyzcjy
fa2f677e18 Fix torch memory saver not enabled in DP scenario (#5560) 2025-04-20 14:20:52 -07:00
Lianmin Zheng
5493c3343e Fix data parallel + tensor parallel (#4499) 2025-03-17 05:13:16 -07:00
Lianmin Zheng
8e66fbecee Improve DP attention (#4390)
Co-authored-by: dhou-xai <dhou@x.ai>
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
2025-03-13 08:23:56 -07:00
Lianmin Zheng
66301e124f Improve code styles (#4021) 2025-03-03 03:20:23 -08:00