{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T19:55:12Z","timestamp":1790884512586,"version":"4.1.0"},"reference-count":38,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Science and Technology Project of the State Grid Corporation of China","award":["522700250006-135-ZN"],"award-info":[{"award-number":["522700250006-135-ZN"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2026,11]]},"DOI":"10.1109\/tpds.2026.3732699","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T19:06:00Z","timestamp":1788980760000},"page":"2524-2538","source":"Crossref","is-referenced-by-count":0,"title":["EcoxPU: Autonomous Multi-Agent Orchestration for Energy-Efficient Disaggregated LLM Inference on Heterogeneous xPUs"],"prefix":"10.1109","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2042-2123","authenticated-orcid":false,"given":"Dong","family":"Dong","sequence":"first","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengzhang","family":"Wu","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-8342-9504","authenticated-orcid":false,"given":"Zheng","family":"Chen","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Longchuan","family":"Yan","sequence":"additional","affiliation":[{"name":"State Grid Information & Telecommunication Center","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianing","family":"Niu","sequence":"additional","affiliation":[{"name":"State Grid Information & Telecommunication Center","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Yu","sequence":"additional","affiliation":[{"name":"State Grid Electric Power Research Institute","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhicheng","family":"Ma","sequence":"additional","affiliation":[{"name":"State Grid Gansu Power Company","place":["Lanzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7656-6428","authenticated-orcid":false,"given":"Jidong","family":"Zhai","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Deepseek-v3 technical report","author":"Liu","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3632775.3662830"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00102"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3727200.3727217"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00103"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ASAP65064.2025.00031"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651380"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2024.3406038"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3669940.3707239"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00113"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3715996"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731092"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CLUSTER59342.2025.11186463"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651352"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3656507"},{"key":"ref16","article-title":"Megatron-LM: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref18","first-page":"288","article-title":"MegaBlocks: Efficient sparse training with mixture-of-experts","volume-title":"Proc. Mach. Learn. Syst.","volume":"5","author":"Gale","year":"2023"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2024.3385639"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3732941"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3773772"},{"key":"ref22","first-page":"193","article-title":"DistServe: Disaggregating prefill and decoding for goodput-optimized large language model serving","volume-title":"Proc. 18th USENIX Symp. Operating Syst. Des. Implementation","author":"Zhong","year":"2024"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2000"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589057"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00019"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750506"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3655956"},{"key":"ref29","article-title":"VoltanaLLM: Feedback-driven frequency control and state-space routing for energy-efficient LLM serving","author":"Yu","year":"2025"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3732941"},{"key":"ref31","article-title":"xdeepserve: Model-as-a-service on Huawei CloudMatrix384","author":"Xiao","year":"2025"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3716025"},{"key":"ref33","article-title":"EcoServe: Enabling cost-effective LLM serving with proactive intra-and inter-instance orchestration","author":"Du","year":"2025"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3447545.3451192"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2022.3144614"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICUAS65942.2025.11007927"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10611573"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.ipm.2026.104878"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/71\/11689667\/11684983.pdf?arnumber=11684983","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T19:03:50Z","timestamp":1790881430000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11684983\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,11]]},"references-count":38,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2026.3732699","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"value":"1045-9219","type":"print"},{"value":"1558-2183","type":"electronic"},{"value":"2161-9883","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,11]]}}}