{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,6]],"date-time":"2026-02-06T11:46:15Z","timestamp":1770378375864,"version":"3.49.0"},"reference-count":54,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","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":[{"name":"National Key Research and Development Program of China","award":["2024YFB2906700"],"award-info":[{"award-number":["2024YFB2906700"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62325205"],"award-info":[{"award-number":["62325205"]}],"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":["U25B2035"],"award-info":[{"award-number":["U25B2035"]}],"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":["62502194"],"award-info":[{"award-number":["62502194"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004608","name":"Natural Science Foundation of Jiangsu Province","doi-asserted-by":"publisher","award":["BK20243053"],"award-info":[{"award-number":["BK20243053"]}],"id":[{"id":"10.13039\/501100004608","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Nanjing University\u2013China Mobile Communications Group Company Ltd","award":["BK20243053"],"award-info":[{"award-number":["BK20243053"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Netw."],"published-print":{"date-parts":[[2026]]},"DOI":"10.1109\/ton.2026.3658194","type":"journal-article","created":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T20:58:44Z","timestamp":1769633924000},"page":"3061-3074","source":"Crossref","is-referenced-by-count":0,"title":["Rail: ReArranging Inter-GPU Links for GPU-Centric Clusters"],"prefix":"10.1109","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-8741-0625","authenticated-orcid":false,"given":"Haixin","family":"Nan","sequence":"first","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Jun","family":"Xu","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0222-4943","authenticated-orcid":false,"given":"Peirui","family":"Cao","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5725-3661","authenticated-orcid":false,"given":"Zhaochen","family":"Zhang","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Yizhi","family":"Wang","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Zhehao","family":"Lin","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Yuhang","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Chengyuan","family":"Huang","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"given":"Wentao","family":"Fan","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"given":"Xiaohu","family":"Xu","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9732-6338","authenticated-orcid":false,"given":"Zhongming","family":"Ji","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"given":"Shengju","family":"Zhang","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"given":"Dongxu","family":"Wang","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9905-6328","authenticated-orcid":false,"given":"Lingkun","family":"Meng","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Company Ltd., Suzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1565-9997","authenticated-orcid":false,"given":"Rong","family":"Gu","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6934-1685","authenticated-orcid":false,"given":"Guihai","family":"Chen","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2710-7628","authenticated-orcid":false,"given":"Chen","family":"Tian","sequence":"additional","affiliation":[{"name":"State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China"}]}],"member":"263","reference":[{"issue":"120","key":"ref1","first-page":"1","article-title":"Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity","volume":"23","author":"Fedus","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref2","first-page":"785","article-title":"Scaling distributed machine learning with in-network aggregation","volume-title":"Proc. 18th USENIX Symp. Networked Syst. Design Implement. (NSDI)","author":"Sapio"},{"key":"ref3","first-page":"829","article-title":"In-network aggregation for shared machine learning clusters","volume-title":"Proc. Mach. Learn. Syst.","volume":"3","author":"Gebara"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3437963.3441727"},{"key":"ref5","volume-title":"NVlink High-Speed GPU Interconnect","year":"2023"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC19947.2020.9063103"},{"key":"ref7","article-title":"Rail-only: A low-cost high-performance network for training LLMs with trillion parameters","author":"Wang","year":"2023","journal-title":"arXiv:2307.12169"},{"key":"ref8","volume-title":"Performance Reported By NCCL Tests","year":"2025"},{"key":"ref9","volume-title":"Sequence To Sequence Learning With Neural Networks: What a Decade","author":"Sutskever","year":"2024"},{"key":"ref10","article-title":"DeepSeek-V3 technical report","volume-title":"arXiv:2412.19437","author":"Liu","year":"2024"},{"key":"ref11","first-page":"593","article-title":"TACCL: Guiding collective algorithm synthesis using communication sketches","volume-title":"Proc. 20th USENIX Symp. Networked Syst. Design Implement. (NSDI)","author":"Shah"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2019.2928289"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3341302.3342085"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3406591"},{"key":"ref15","first-page":"131","article-title":"Revisiting congestion control for lossless Ethernet","volume-title":"Proc. 21st USENIX Symp. Networked Syst. Design Implement. (NSDI)","author":"Zhang"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527382"},{"key":"ref17","volume-title":"NCCL","year":"2021"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651362"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672239"},{"key":"ref20","volume-title":"Doubling All2all Performance With Nvidia Collective Communication Library 2.12","year":"2022"},{"key":"ref21","volume-title":"NS-3","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/s11023-020-09548-1"},{"issue":"240","key":"ref23","first-page":"1","article-title":"PaLM: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref24","article-title":"Scaling laws for neural language models","author":"Kaplan","year":"2020","journal-title":"arXiv:2001.08361"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3405671.3405810"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2024.3360081"},{"key":"ref27","volume-title":"Nvswitch: Leveraging Nvlink to Maximum Effect","year":"2018"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00071"},{"key":"ref29","volume-title":"Nvidia L20 Enterprise 48Gb","year":"2023"},{"key":"ref30","volume-title":"Nvidia L40s","year":"2023"},{"key":"ref31","volume-title":"Nvidia A800 40Gb Active Graphics Card","year":"2023"},{"key":"ref32","volume-title":"Nvidia Tesla H800","year":"2024"},{"key":"ref33","volume-title":"Nvidia Tesla V100","year":"2024"},{"key":"ref34","first-page":"745","article-title":"MegaScale: Scaling large language model training to more than 10,000 GPUs","volume-title":"Proc. 21st USENIX Symp. Networked Syst. Design Implement. (NSDI)","author":"Jiang"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750521"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672265"},{"key":"ref37","first-page":"1","article-title":"Efficient large-scale language model training on GPU clusters using megatron-LM","volume-title":"Proc. Int. Conf. High Perform. Comput., Netw., Storage Anal.","author":"Narayanan"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2008.09.002"},{"key":"ref40","volume-title":"NVIDIA H100 Tensor Core GPU","year":"2024"},{"key":"ref41","volume-title":"Connectx-7 400G Adapters","year":"2024"},{"key":"ref42","volume-title":"Production Brief for Intel\u00ae Ethernet Controller E810-CAM2\/CAM1\/XXVAM2","year":"2020"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3627703.3629574"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/90.769767"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/1592681.1592693"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3131346"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2025.3592688"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1145\/3437801.3441620"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672249"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750499"},{"key":"ref51","article-title":"ForestColl: Throughput-optimal collective communications on heterogeneous network fabrics","author":"Zhao","year":"2024","journal-title":"arXiv:2402.06787"},{"key":"ref52","article-title":"PAT: A new algorithm for all-gather and reduce-scatter operations at scale","author":"Jeaugey","year":"2025","journal-title":"arXiv:2506.20252"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1147\/JRD.2019.2947013"},{"key":"ref54","article-title":"Collective communication for 100k+ GPUs","author":"Si","year":"2025","journal-title":"arXiv:2510.20171"}],"container-title":["IEEE Transactions on Networking"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10723154\/11317935\/11364238.pdf?arnumber=11364238","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T20:39:32Z","timestamp":1770323972000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11364238\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":54,"URL":"https:\/\/doi.org\/10.1109\/ton.2026.3658194","relation":{},"ISSN":["2998-4157"],"issn-type":[{"value":"2998-4157","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}