{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T02:03:15Z","timestamp":1780020195151,"version":"3.53.1"},"reference-count":38,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/100014103","name":"Key Technology Research and Development Program of Shandong Province","doi-asserted-by":"publisher","award":["2025CXPT098"],"award-info":[{"award-number":["2025CXPT098"]}],"id":[{"id":"10.13039\/100014103","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100016109","name":"Taishan Industry Leading Talents","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100016109","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100007129","name":"Natural Science Foundation of Shandong Province","doi-asserted-by":"publisher","award":["ZR2022QF030"],"award-info":[{"award-number":["ZR2022QF030"]}],"id":[{"id":"10.13039\/501100007129","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Future Generation Computer Systems"],"published-print":{"date-parts":[[2026,9]]},"DOI":"10.1016\/j.future.2026.108459","type":"journal-article","created":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T16:25:58Z","timestamp":1772555158000},"page":"108459","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["A reinforcement learning-based approach for scheduling ML training tasks in heterogeneous Kubernetes clusters"],"prefix":"10.1016","volume":"182","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-0781-3201","authenticated-orcid":false,"given":"Shuwei","family":"Dong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bingbing","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6157-3740","authenticated-orcid":false,"given":"Li","family":"Pan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4108-1391","authenticated-orcid":false,"given":"Shijun","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"1","key":"10.1016\/j.future.2026.108459_bib0001","doi-asserted-by":"crossref","first-page":"381","DOI":"10.21275\/ART20203995","article-title":"Machine learning algorithms-a review","volume":"9","author":"Mahesh","year":"2020","journal-title":"Int. J. Sci. Res. (IJSR).[Internet]"},{"key":"10.1016\/j.future.2026.108459_bib0002","series-title":"19Th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)","first-page":"945","article-title":"MLaaS in the wild: workload analysis and scheduling in large-Scale heterogeneous GPU clusters","author":"Weng","year":"2022"},{"key":"10.1016\/j.future.2026.108459_bib0003","series-title":"2012 International Conference on Computing, Electronics and Electrical Technologies (ICCEET)","first-page":"877","article-title":"Cloud computing-concepts, architecture and challenges","author":"Jadeja","year":"2012"},{"issue":"3","key":"10.1016\/j.future.2026.108459_bib0004","doi-asserted-by":"crossref","first-page":"677","DOI":"10.1109\/TCC.2017.2702586","article-title":"Cloud container technologies: a state-of-the-art review","volume":"7","author":"Pahl","year":"2017","journal-title":"IEEE Trans. Cloud Comput."},{"key":"10.1016\/j.future.2026.108459_bib0005","unstructured":"Kubernetes, 2024, https:\/\/kubernetes.io."},{"issue":"7","key":"10.1016\/j.future.2026.108459_bib0006","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3539606","article-title":"Kubernetes scheduling: taxonomy, ongoing issues and challenges","volume":"55","author":"Carri\u00f3n","year":"2022","journal-title":"ACM Comput. Surv."},{"issue":"2","key":"10.1016\/j.future.2026.108459_bib0007","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1109\/MM.2010.41","article-title":"The GPU computing era","volume":"30","author":"Nickolls","year":"2010","journal-title":"IEEE Micro."},{"key":"10.1016\/j.future.2026.108459_bib0008","series-title":"2012 Proceedings IEEE Infocom","first-page":"702","article-title":"Stochastic models of load balancing and scheduling in cloud computing clusters","author":"Maguluri","year":"2012"},{"key":"10.1016\/j.future.2026.108459_bib0009","series-title":"2013 22Nd Australian Software Engineering Conference","first-page":"78","article-title":"An empirical investigation on the simulation of priority and shortest-job-first scheduling for cloud-based software systems","author":"Ru","year":"2013"},{"key":"10.1016\/j.future.2026.108459_bib0010","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1016\/j.jnca.2017.04.007","article-title":"Load-balancing algorithms in cloud computing: a survey","volume":"88","author":"Ghomi","year":"2017","journal-title":"J. Network Comput. Appl."},{"issue":"6","key":"10.1016\/j.future.2026.108459_bib0011","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1109\/MSP.2017.2743240","article-title":"Deep reinforcement learning: a brief survey","volume":"34","author":"Arulkumaran","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"10.1016\/j.future.2026.108459_bib0012","series-title":"2018 23rd Asia and South Pacific Design Automation Conference (ASP-DAC)","first-page":"129","article-title":"DRL-cloud: deep reinforcement learning-based resource provisioning and task scheduling for cloud service providers","author":"Cheng","year":"2018"},{"issue":"2","key":"10.1016\/j.future.2026.108459_bib0013","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1109\/TC.2020.2987567","article-title":"A3C-DO: A regional resource scheduling framework based on deep reinforcement learning in edge scenario","volume":"70","author":"Zou","year":"2020","journal-title":"IEEE Trans. Comput."},{"key":"10.1016\/j.future.2026.108459_bib0014","series-title":"Proceedings of the 18th Annual ACM\/SIGDA International Symposium on Field Programmable Gate Arrays","first-page":"115","article-title":"Axel: a heterogeneous cluster with FPGAs and GPUs","author":"Tsoi","year":"2010"},{"key":"10.1016\/j.future.2026.108459_bib0015","series-title":"NOMS 2020-2020 IEEE\/IFIP Network Operations and Management Symposium","first-page":"1","article-title":"Adaptive scaling of kubernetes pods","author":"Balla","year":"2020"},{"key":"10.1016\/j.future.2026.108459_bib0016","series-title":"Proceedings of the 5th European Conference on Computer Systems","first-page":"237","article-title":"Q-clouds: managing performance interference effects for qos-aware clouds","author":"Nathuji","year":"2010"},{"key":"10.1016\/j.future.2026.108459_bib0017","series-title":"Proceedings of the 29th International Symposium on High-performance Parallel and Distributed Computing","first-page":"173","article-title":"Kubeshare: a framework to manage GPUs as first-class and shared resources in container cloud","author":"Yeh","year":"2020"},{"key":"10.1016\/j.future.2026.108459_bib0018","unstructured":"Alibaba, gpushare device plugin, 2024, https:\/\/github.com\/AliyunContainerService\/gpushare-device-plugin."},{"key":"10.1016\/j.future.2026.108459_bib0019","unstructured":"Nvidia, Nvidia GPU device plugin, 2024, https:\/\/github.com\/NVIDIA\/k8s-device-plugin."},{"key":"10.1016\/j.future.2026.108459_bib0020","unstructured":"V. Mnih, Playing atari with deep reinforcement learning, (2013). arXiv: 1312.5602."},{"key":"10.1016\/j.future.2026.108459_bib0021","series-title":"IEEE INFOCOM 2019-IEEE Conference on Computer Communications","first-page":"505","article-title":"Deep learning-based job placement in distributed machine learning clusters","author":"Bao","year":"2019"},{"key":"10.1016\/j.future.2026.108459_bib0022","series-title":"International Conference on Machine Learning","first-page":"8280","article-title":"Off-policy reinforcement learning with delayed rewards","author":"Han","year":"2022"},{"key":"10.1016\/j.future.2026.108459_bib0023","unstructured":"S. Huang, S. Onta\u00f1\u00f3n, A closer look at invalid action masking in policy gradient algorithms, (2020). arXiv: 2006.14171."},{"key":"10.1016\/j.future.2026.108459_bib0024","unstructured":"Tensorflow, 2024, https:\/\/tensorflow.google.cn\/tutorials."},{"issue":"2","key":"10.1016\/j.future.2026.108459_bib0025","first-page":"213","article-title":"KubCG: a dynamic kubernetes scheduler for heterogeneous clusters","volume":"51","author":"El Haj Ahmed","year":"2021","journal-title":"Software: Practice Exp."},{"issue":"10","key":"10.1016\/j.future.2026.108459_bib0026","first-page":"2102","article-title":"DRS: A deep reinforcement learning enhanced kubernetes scheduler for microservice-based system","volume":"54","author":"Jian","year":"2024","journal-title":"Software: Practice Exp."},{"key":"10.1016\/j.future.2026.108459_bib0027","unstructured":"D.P. Kingma, Adam: A method for stochastic optimization, (2014). arXiv: 1412.6980."},{"key":"10.1016\/j.future.2026.108459_bib0028","series-title":"14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20)","first-page":"481","article-title":"{Heterogeneity-Aware} Cluster scheduling policies for deep learning workloads","author":"Narayanan","year":"2020"},{"key":"10.1016\/j.future.2026.108459_bib0029","series-title":"2020 IEEE International Conference on Cloud Engineering (IC2E)","first-page":"116","article-title":"Rlsk: a job scheduler for federated kubernetes clusters based on reinforcement learning","author":"Huang","year":"2020"},{"issue":"5","key":"10.1016\/j.future.2026.108459_bib0030","doi-asserted-by":"crossref","first-page":"4267","DOI":"10.1007\/s11227-020-03427-3","article-title":"KCSS: Kubernetes container scheduling strategy","volume":"77","author":"Menouer","year":"2021","journal-title":"J. Supercomput."},{"key":"10.1016\/j.future.2026.108459_bib0031","series-title":"Economics of Grids, Clouds, Systems, and Services: 14th International Conference, GECON 2017, Biarritz, France, September 19-21, 2017, Proceedings 14","first-page":"162","article-title":"Client-side scheduling based on application characterization on kubernetes","author":"Medel","year":"2017"},{"key":"10.1016\/j.future.2026.108459_bib0032","series-title":"2018 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Ubiquitous Computing & Communications, Big Data & Cloud Computing, Social Computing & Networking, Sustainable Computing & Communications (ISPA\/IUCC\/BDCloud\/SocialCom\/SustainCom)","first-page":"252","article-title":"Gaia scheduler: a kubernetes-based scheduler framework","author":"Song","year":"2018"},{"key":"10.1016\/j.future.2026.108459_bib0033","series-title":"IEEE INFOCOM 2018-IEEE Conference on Computer Communications","first-page":"495","article-title":"Online job scheduling in distributed machine learning clusters","author":"Bao","year":"2018"},{"key":"10.1016\/j.future.2026.108459_bib0034","series-title":"2017 IEEE International Conference on Smart Computing (SMARTCOMP)","first-page":"1","article-title":"Towards distributed machine learning in shared clusters: a dynamically-partitioned approach","author":"Sun","year":"2017"},{"key":"10.1016\/j.future.2026.108459_bib0035","series-title":"2023 USENIX Annual Technical Conference (USENIX ATC 23)","first-page":"995","article-title":"Beware of fragmentation: scheduling GPU-Sharing workloads with fragmentation gradient descent","author":"Weng","year":"2023"},{"key":"10.1016\/j.future.2026.108459_bib0036","series-title":"Closer","first-page":"569","article-title":"DRAGON: A dynamic scheduling and scaling controller for managing distributed deep learning jobs in kubernetes cluster","author":"Lin","year":"2019"},{"issue":"3","key":"10.1016\/j.future.2026.108459_bib0037","doi-asserted-by":"crossref","first-page":"3770","DOI":"10.1109\/JSYST.2021.3129974","article-title":"Speculative container scheduling for deep learning applications in a kubernetes cluster","volume":"16","author":"Mao","year":"2021","journal-title":"IEEE Syst. J."},{"issue":"11","key":"10.1016\/j.future.2026.108459_bib0038","first-page":"2968","article-title":"Microservice deployment in edge computing based on deep q learning","volume":"33","author":"Lv","year":"2022","journal-title":"IEEE Trans. Parallel Distrib. Syst."}],"container-title":["Future Generation Computer Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0167739X26000932?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0167739X26000932?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T01:34:39Z","timestamp":1780018479000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0167739X26000932"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":38,"alternative-id":["S0167739X26000932"],"URL":"https:\/\/doi.org\/10.1016\/j.future.2026.108459","relation":{},"ISSN":["0167-739X"],"issn-type":[{"value":"0167-739X","type":"print"}],"subject":[],"published":{"date-parts":[[2026,9]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"A reinforcement learning-based approach for scheduling ML training tasks in heterogeneous Kubernetes clusters","name":"articletitle","label":"Article Title"},{"value":"Future Generation Computer Systems","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.future.2026.108459","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"108459"}}