{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T18:00:32Z","timestamp":1785952832059,"version":"3.56.0"},"reference-count":48,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62302126"],"award-info":[{"award-number":["62302126"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62302123"],"award-info":[{"award-number":["62302123"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Shenzhen Science and Technology Program","award":["RCBS20221008093125065"],"award-info":[{"award-number":["RCBS20221008093125065"]}]},{"name":"Shenzhen Science and Technology Program","award":["KJZD20240903104103005"],"award-info":[{"award-number":["KJZD20240903104103005"]}]},{"name":"Shenzhen Science and Technology Program","award":["KJZD20230923114213027"],"award-info":[{"award-number":["KJZD20230923114213027"]}]},{"name":"Shenzhen Science and Technology Program","award":["KJZD20230923115113026"],"award-info":[{"award-number":["KJZD20230923115113026"]}]},{"name":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies","award":["2022B1212010005"],"award-info":[{"award-number":["2022B1212010005"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Cloud Comput."],"published-print":{"date-parts":[[2026,1]]},"DOI":"10.1109\/tcc.2025.3624752","type":"journal-article","created":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T17:59:47Z","timestamp":1761242387000},"page":"45-61","source":"Crossref","is-referenced-by-count":1,"title":["Castor: Optimizing Deep Learning Job Scheduling in Multi-Tenant GPU Clusters via Intelligent Colocation"],"prefix":"10.1109","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-0331-4278","authenticated-orcid":false,"given":"Yizhou","family":"Luo","sequence":"first","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiaxin","family":"Lai","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1418-5160","authenticated-orcid":false,"given":"Shaohuai","family":"Shi","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9480-5632","authenticated-orcid":false,"given":"Chen","family":"Chen","sequence":"additional","affiliation":[{"name":"John Hopcroft Center for Computer Science, Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6903-145X","authenticated-orcid":false,"given":"Shuhan","family":"Qi","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6611-2046","authenticated-orcid":false,"given":"Jiajia","family":"Zhang","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2986-967X","authenticated-orcid":false,"given":"Qiang","family":"Wang","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.112.2100706"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2025.3553922"},{"key":"ref4","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Neural Inf. Process. Syst.","author":"Brown","year":"2020"},{"key":"ref5","first-page":"1223","article-title":"Large scale distributed deep networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"25","author":"Dean","year":"2012"},{"key":"ref6","first-page":"947","article-title":"Analysis of large-scale multi-tenant GPU clusters for DNN training workloads","volume-title":"Proc. USENIX Annu. Tech. Conf.","author":"Jeon","year":"2019"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/2785956.2787488"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737415"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/2901318.2901355"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737465"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00204"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2017.8057205"},{"key":"ref13","first-page":"27","article-title":"Low-latency job scheduling with preemption for the development of deep learning","volume-title":"Proc. USENIX Conf. Oper. Mach. Learn.","author":"Yabuuchi","year":"2019"},{"key":"ref14","first-page":"39","article-title":"Tony: An orchestrator for distributed machine learning jobs","volume-title":"Proc. 2019 USENIX Conf. Oper. Mach. Learn.","author":"Hsu","year":"2019"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3190508.3190517"},{"key":"ref16","first-page":"485","article-title":"Tiresias: A GPU cluster manager for distributed deep learning","volume-title":"Proc. 16th USENIX Symp. Networked Syst. Des. Implementation","author":"Gu","year":"2019"},{"key":"ref17","first-page":"481","article-title":"Heterogeneity-aware cluster scheduling policies for deep learning workloads","volume-title":"Proc. 14th USENIX Symp. Operating Syst. Des. Implementation","author":"Narayanan","year":"2020"},{"key":"ref18","first-page":"1","article-title":"Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning","volume-title":"Proc. 15th USENIX Symp. Operating Syst. Des. Implementation","author":"Qiao","year":"2021"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3627703.3629583"},{"key":"ref20","first-page":"595","article-title":"Gandiva: Introspective cluster scheduling for deep learning","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Des. Implementation","author":"Xiao","year":"2018"},{"key":"ref21","first-page":"161","article-title":"Zico: Efficient GPU memory sharing for concurrent DNN training","volume-title":"Proc. 2021 USENIX Annu. Tech. Conf.","author":"Lim","year":"2021"},{"key":"ref22","first-page":"98","article-title":"Salus: Fine-grained GPU sharing primitives for deep learning applications","volume-title":"Proc. 3rd MLSys Conf.","volume":"2","author":"Yu","year":"2019"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575705"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid57682.2023.00040"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.48550\/arxiv.1811.06965"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359646"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/IWQoS61813.2024.10682877"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476223"},{"key":"ref29","first-page":"947","article-title":"Analysis of large-scale multi-tenant GPU clusters for DNN training workloads","volume-title":"Proc. 2019 USENIX Annu. Tech. Conf.","author":"Jeon","year":"2019"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3337821.3337873"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737460"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00201"},{"key":"ref33","first-page":"721","article-title":"Elastic resource sharing for distributed deep learning","volume-title":"Proc. USENIX Assoc.","author":"Hwang","year":"2021"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640375"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613175"},{"key":"ref36","first-page":"945","article-title":"MLaaS in the wild: Workload analysis and scheduling in large-scale heterogeneous GPU clusters","volume-title":"Proc. 19th USENIX Symp. Netw. Syst. Des. Implementation","author":"Weng","year":"2022"},{"key":"ref37","article-title":"Communication contention aware scheduling of multiple deep learning training jobs","author":"Wang","year":"2020"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM48880.2022.9796785"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3492866.3549716"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-24685-5_1"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1177\/1094342005051521"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-15646-5_3"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378530"},{"key":"ref44","first-page":"289","article-title":"Themis: Fair and efficient GPU cluster scheduling","volume-title":"Proc. 17th USENIX Symp. Networked Syst. Des. Implementation","author":"Mahajan","year":"2020"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1016\/j.simpat.2010.08.010"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i11.29091"},{"key":"ref47","article-title":"Nvidia DCGM","year":"2023"},{"key":"ref48","article-title":"Nvidia multi-process service","year":"2023"}],"container-title":["IEEE Transactions on Cloud Computing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6245519\/11427054\/11216001.pdf?arnumber=11216001","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T05:21:15Z","timestamp":1773120075000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11216001\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":48,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tcc.2025.3624752","relation":{},"ISSN":["2168-7161","2372-0018"],"issn-type":[{"value":"2168-7161","type":"electronic"},{"value":"2372-0018","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]}}}