{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T05:02:15Z","timestamp":1784782935438,"version":"3.55.0"},"reference-count":24,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100014188","name":"Ministry of Science and ICT, South Korea","doi-asserted-by":"publisher","award":["RS-2025-02220502"],"award-info":[{"award-number":["RS-2025-02220502"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100010418","name":"Institute of Information & Communications Technology Planning & Evaluation","doi-asserted-by":"publisher","award":["IITP-2026-RS-2022-00156439"],"award-info":[{"award-number":["IITP-2026-RS-2022-00156439"]}],"id":[{"id":"10.13039\/501100010418","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":[[2026,12]]},"DOI":"10.1016\/j.future.2026.108659","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T23:59:18Z","timestamp":1781135958000},"page":"108659","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["REX: Adaptive distributed training with statistically validated batch scaling and spline-guided Bayesian optimization on heterogeneous GPU clusters"],"prefix":"10.1016","volume":"185","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-5078-4405","authenticated-orcid":false,"given":"HyungJun","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hwamin","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Heonchang","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.future.2026.108659_b1","series-title":"2025 IEEE 18th International Conference on Cloud Computing","first-page":"409","article-title":"HEART: Heterogeneous-aware traffic allocation in multi-replica deployments on kubernetes","author":"Park","year":"2025"},{"issue":"4","key":"10.1016\/j.future.2026.108659_b2","doi-asserted-by":"crossref","first-page":"1217","DOI":"10.1109\/TETCI.2022.3220224","article-title":"DLB: a dynamic load balance strategy for distributed training of deep neural networks","volume":"7","author":"Ye","year":"2022","journal-title":"IEEE Trans. Emerg. Top. Comput. Intell."},{"key":"10.1016\/j.future.2026.108659_b3","series-title":"15th USENIX Symposium on Operating Systems Design and Implementation","first-page":"1","article-title":"Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning","author":"Qiao","year":"2021"},{"key":"10.1016\/j.future.2026.108659_b4","doi-asserted-by":"crossref","unstructured":"S. Jayaram Subramanya, D. Arfeen, S. Lin, A. Qiao, Z. Jia, G.R. Ganger, Sia: Heterogeneity-aware, goodput-optimized ML-cluster scheduling, in: Proceedings of the 29th Symposium on Operating Systems Principles, 2023, pp. 642\u2013657.","DOI":"10.1145\/3600006.3613175"},{"key":"10.1016\/j.future.2026.108659_b5","doi-asserted-by":"crossref","unstructured":"Z. Mo, H. Xu, C. Xu, Heet: Accelerating elastic training in heterogeneous deep learning clusters, in: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Vol. 2, 2024, pp. 499\u2013513.","DOI":"10.1145\/3620665.3640375"},{"key":"10.1016\/j.future.2026.108659_b6","article-title":"Practical bayesian optimization of machine learning algorithms","volume":"25","author":"Snoek","year":"2012","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.future.2026.108659_b7","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1214\/aoms\/1177730491","article-title":"On a test of whether one of two random variables is stochastically larger than the other","author":"Mann","year":"1947","journal-title":"Ann. Math. Stat."},{"key":"10.1016\/j.future.2026.108659_b8","series-title":"AdaptDL standalone runtime (documentation)","author":"Petuum","year":"2025"},{"key":"10.1016\/j.future.2026.108659_b9","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.future.2026.108659_b10","article-title":"More effective distributed ml via a stale synchronous parallel parameter server","volume":"26","author":"Ho","year":"2013","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.future.2026.108659_b11","series-title":"2019 IEEE 39th International Conference on Distributed Computing Systems","first-page":"1507","article-title":"Dynamic stale synchronous parallel distributed training for deep learning","author":"Zhao","year":"2019"},{"issue":"4","key":"10.1016\/j.future.2026.108659_b12","doi-asserted-by":"crossref","first-page":"2637","DOI":"10.1109\/TCC.2021.3062398","article-title":"GSSP: Eliminating stragglers through grouping synchronous for distributed deep learning in heterogeneous cluster","volume":"10","author":"Sun","year":"2021","journal-title":"IEEE Trans. Cloud Comput."},{"issue":"5","key":"10.1016\/j.future.2026.108659_b13","doi-asserted-by":"crossref","first-page":"1030","DOI":"10.1109\/TPDS.2020.3040601","article-title":"Petrel: Heterogeneity-aware distributed deep learning via hybrid synchronization","volume":"32","author":"Zhou","year":"2020","journal-title":"IEEE Trans. Parallel Distrib. Syst."},{"key":"10.1016\/j.future.2026.108659_b14","series-title":"2021 IEEE 41st International Conference on Distributed Computing Systems","first-page":"528","article-title":"Sync-switch: Hybrid parameter synchronization for distributed deep learning","author":"Li","year":"2021"},{"key":"10.1016\/j.future.2026.108659_b15","doi-asserted-by":"crossref","unstructured":"Q. Luo, J. He, Y. Zhuo, X. Qian, Prague: High-performance heterogeneity-aware asynchronous decentralized training, in: Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems, 2020, pp. 401\u2013416.","DOI":"10.1145\/3373376.3378499"},{"key":"10.1016\/j.future.2026.108659_b16","unstructured":"L. Mai, G. Li, M. Wagenl\u00e4nder, K. Fertakis, A.-O. Brabete, P. Pietzuch, {KungFu}: Making training in distributed machine learning adaptive, in: 14th USENIX Symposium on Operating Systems Design and Implementation, OSDI 20, 2020, pp. 937\u2013954."},{"key":"10.1016\/j.future.2026.108659_b17","first-page":"1","article-title":"LBB: load-balanced batching for efficient distributed learning on heterogeneous GPU cluster","author":"Yao","year":"2024","journal-title":"J. Supercomput."},{"key":"10.1016\/j.future.2026.108659_b18","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1007\/s11227-019-02845-2","article-title":"BOA: batch orchestration algorithm for straggler mitigation of distributed DL training in heterogeneous GPU cluster","volume":"76","author":"Yang","year":"2020","journal-title":"J. Supercomput."},{"key":"10.1016\/j.future.2026.108659_b19","doi-asserted-by":"crossref","unstructured":"T.N. Le, X. Sun, M. Chowdhury, Z. Liu, AlloX: Compute allocation in hybrid clusters, in: Proceedings of the Fifteenth European Conference on Computer Systems, 2020, pp. 1\u201316.","DOI":"10.1145\/3342195.3387547"},{"key":"10.1016\/j.future.2026.108659_b20","unstructured":"D. Narayanan, K. Santhanam, F. Kazhamiaka, A. Phanishayee, M. Zaharia, {Heterogeneity-Aware} cluster scheduling policies for deep learning workloads, in: 14th USENIX Symposium on Operating Systems Design and Implementation, OSDI 20, 2020, pp. 481\u2013498."},{"key":"10.1016\/j.future.2026.108659_b21","unstructured":"Z. Xu, M. Yu, F.Y. Yan, Decouple and Decompose: Scaling Resource Allocation with {DeDe}, in: 19th USENIX Symposium on Operating Systems Design and Implementation, OSDI 25, 2025, pp. 393\u2013409."},{"key":"10.1016\/j.future.2026.108659_b22","doi-asserted-by":"crossref","DOI":"10.1016\/j.future.2025.107883","article-title":"AdaGap: An adaptive gap-aware resource allocation strategy for GPU sharing in heterogeneous clusters","author":"Wang","year":"2025","journal-title":"Future Gener. Comput. Syst."},{"key":"10.1016\/j.future.2026.108659_b23","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1016\/j.future.2024.07.001","article-title":"PowerTrain: Fast, generalizable time and power prediction models to optimize DNN training on accelerated edges","volume":"161","author":"Prashanthi","year":"2024","journal-title":"Future Gener. Comput. Syst."},{"key":"10.1016\/j.future.2026.108659_b24","doi-asserted-by":"crossref","DOI":"10.1016\/j.sysarc.2025.103573","article-title":"SADDLE: A runtime feedback control architecture for adaptive distributed deep learning in heterogeneous GPU clusters","author":"Kim","year":"2025","journal-title":"J. Syst. Archit."}],"container-title":["Future Generation Computer Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0167739X26002931?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0167739X26002931?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T04:21:38Z","timestamp":1784780498000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0167739X26002931"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,12]]},"references-count":24,"alternative-id":["S0167739X26002931"],"URL":"https:\/\/doi.org\/10.1016\/j.future.2026.108659","relation":{},"ISSN":["0167-739X"],"issn-type":[{"value":"0167-739X","type":"print"}],"subject":[],"published":{"date-parts":[[2026,12]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"REX: Adaptive distributed training with statistically validated batch scaling and spline-guided Bayesian optimization on heterogeneous GPU clusters","name":"articletitle","label":"Article Title"},{"value":"Future Generation Computer Systems","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.future.2026.108659","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":"108659"}}