{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:44:32Z","timestamp":1782798272027,"version":"3.54.5"},"reference-count":37,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"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","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","id":[{"id":"10.13039\/501100004608","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,5,18]]},"DOI":"10.1109\/infocom59046.2026.11571222","type":"proceedings-article","created":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T19:38:15Z","timestamp":1782761895000},"page":"1-10","source":"Crossref","is-referenced-by-count":0,"title":["Chameleon: Adaptive Fault Tolerance for Distributed Training via Real-time Policy Selection"],"prefix":"10.1109","author":[{"given":"Yuhang","family":"Zhou","sequence":"first","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhibin","family":"Wang","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peng","family":"Jiang","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haoran","family":"Xia","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junhe","family":"Lu","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qianyu","family":"Jiang","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rong","family":"Gu","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hengxi","family":"Xu","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinjing","family":"Huang","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guanghuan","family":"Fang","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiheng","family":"Hu","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingyi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongjin","family":"Cai","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jian","family":"He","sequence":"additional","affiliation":[{"name":"Huawei,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Tian","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology,China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024"},{"key":"ref2","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613145"},{"key":"ref4","first-page":"929","article-title":"Check-N-Run: a checkpointing system for training deep learning recommendation models","volume-title":"19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)","author":"Eisenman"},{"key":"ref5","article-title":"Megascale: Scaling large language model training to more than 10,000 gpus","author":"Jiang","year":"2024"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS47774.2020.00018"},{"key":"ref7","first-page":"497","article-title":"Bamboo: Making preemptible instances resilient for affordable training of large {DNNs}","volume-title":"20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23)","author":"Thorpe"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613152"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695960"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695975"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00071"},{"key":"ref12","article-title":"Accurate, large minibatch sgd: Training imagenet in 1 hour","author":"Goyal","year":"2018"},{"key":"ref13","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"Sergeev","year":"2018"},{"key":"ref14","doi-asserted-by":"crossref","DOI":"10.1145\/3225058.3225069","article-title":"Imagenet training in minutes","author":"You","year":"2018"},{"key":"ref15","article-title":"Using deepspeed and megatron to train megatron-turing nlg 530b, a large-scale generative language model","author":"Smith","year":"2022"},{"key":"ref16","article-title":"Gpipe: Efficient training of giant neural networks using pipeline parallelism","author":"Huang","year":"2019"},{"key":"ref17","first-page":"16 639","article-title":"BPipe: Memory-balanced pipeline parallelism for training large language models","volume-title":"Proceedings of the 40th International Conference on Machine Learning","volume":"202","author":"Kim"},{"key":"ref18","doi-asserted-by":"crossref","DOI":"10.1109\/SC41405.2020.00024","article-title":"Zero: Memory optimizations toward training trillion parameter models","author":"Rajbhandari","year":"2020"},{"key":"ref19","article-title":"Shortcut-connected expert parallelism for accelerating mixture-of-experts","author":"Cai","year":"2025"},{"key":"ref20","article-title":"Eps-moe: Expert pipeline scheduler for cost-efficient moe inference","author":"Qian","year":"2025"},{"key":"ref21","article-title":"Moe parallel folding: Heterogeneous parallelism mappings for efficient large-scale moe model training with megatron core","author":"Liu","year":"2025"},{"issue":"1","key":"ref22","article-title":"Switch transformers: scaling to trillion parameter models with simple and efficient sparsity","volume":"23","author":"Fedus","year":"2022","journal-title":"J. Mach. Learn. Res"},{"key":"ref23","article-title":"Megascale-moe: Large-scale communication-efficient training of mixture-of-experts models in production","author":"Jin","year":"2025"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3663408.3663409"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS57527.2023.00037"},{"key":"ref26","article-title":"Reducing activation recomputation in large transformer models","author":"Korthikanti","year":"2022"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476145"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3492321.3519584"},{"key":"ref29","article-title":"Elaswave: An elastic-native system for scalable hybrid-parallel training","author":"Kang","year":"2025"},{"key":"ref30","first-page":"1","article-title":"Failures in large scale systems: Long-term measurement, analysis, and implications","volume-title":"SC17: International Conference for High Performance Computing, Networking, Storage and Analysis","author":"Gupta"},{"key":"ref31","first-page":"945","article-title":"MLaaS in the wild: Workload analysis and scheduling in Large-Scale heterogeneous GPU clusters","volume-title":"19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)","author":"Weng"},{"key":"ref32","first-page":"505","article-title":"Minder: Faulty machine detection for large-scale distributed model training","volume-title":"22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25)","author":"Deng"},{"key":"ref33","article-title":"Zero bubble pipeline parallelism","author":"Qi","year":"2023"},{"key":"ref34","first-page":"559","article-title":"Alpa: Automating inter- and Intra-Operator parallelism for distributed deep learning","volume-title":"16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)","author":"Zheng"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1002\/nav.3800020109"},{"key":"ref36","doi-asserted-by":"crossref","DOI":"10.1109\/IPDPSW66978.2025.00216","article-title":"Parallel scan on ascend ai accelerators","author":"Wr\u00f3blewski","year":"2025"},{"key":"ref37","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"}],"event":{"name":"IEEE INFOCOM 2026 - IEEE Conference on Computer Communications","location":"Tokyo, Japan","start":{"date-parts":[[2026,5,18]]},"end":{"date-parts":[[2026,5,21]]}},"container-title":["IEEE INFOCOM 2026 - IEEE Conference on Computer Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11571071\/11571169\/11571222.pdf?arnumber=11571222","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:11:56Z","timestamp":1782796316000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11571222\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1109\/infocom59046.2026.11571222","relation":{},"subject":[],"published":{"date-parts":[[2026,5,18]]}}}