{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T22:13:05Z","timestamp":1784239985428,"version":"3.55.0"},"reference-count":42,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"9","license":[{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"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-037"}],"funder":[{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006132","name":"Office of Science","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006132","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006192","name":"Advanced Scientific Computing Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006192","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Science Foundation","award":["2411386\/2411387"],"award-info":[{"award-number":["2411386\/2411387"]}]},{"name":"National Science Foundation","award":["2106634\/2106635"],"award-info":[{"award-number":["2106634\/2106635"]}]},{"name":"National Science Foundation","award":["2514056"],"award-info":[{"award-number":["2514056"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2026,9]]},"DOI":"10.1109\/tpds.2026.3694405","type":"journal-article","created":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T19:48:09Z","timestamp":1779133689000},"page":"2037-2049","source":"Crossref","is-referenced-by-count":0,"title":["<i>DataStates-LLM<\/i>\n                    : Scalable Checkpointing for Transformer Models Using Composable State Providers"],"prefix":"10.1109","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8200-0148","authenticated-orcid":false,"given":"Avinash","family":"Maurya","sequence":"first","affiliation":[{"name":"Argonne National Laboratory, Lemont, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5034-2880","authenticated-orcid":false,"given":"M. Mustafa","family":"Rafique","sequence":"additional","affiliation":[{"name":"Rochester Institute of Technology, Rochester, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7890-3934","authenticated-orcid":false,"given":"Franck","family":"Cappello","sequence":"additional","affiliation":[{"name":"Argonne National Laboratory, Lemont, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0661-7509","authenticated-orcid":false,"given":"Bogdan","family":"Nicolae","sequence":"additional","affiliation":[{"name":"Argonne National Laboratory, Lemont, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Towards an AI co-scientist","author":"Gottweis","year":"2025"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731410"},{"key":"ref3","article-title":"BLOOM: A 176B-Parameter open-access multilingual language model","author":"Workshop","year":"2023"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.14778\/3611540.3611569"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3731569.3764838"},{"key":"ref7","article-title":"Unicron: Economizing self-healing LLM training at scale","author":"He","year":"2023"},{"key":"ref8","article-title":"Spike no more: Stabilizing the pre-training of large language models","author":"Takase","year":"2023"},{"key":"ref9","article-title":"FastPersist: Accelerating model checkpointing in deep learning","author":"Wang","year":"2024"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3625549.3658685"},{"key":"ref11","article-title":"Welcome to the torchsnapshot documentation","author":"TorchSnapshot","year":"2024"},{"key":"ref12","first-page":"203","article-title":"CheckFreq: Frequent, fine-grained DNN checkpointing","volume-title":"Proc. USENIX Conf. File Storage Technol.","author":"Mohan","year":"2021"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613145"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid49817.2020.00-76"},{"key":"ref15","article-title":"Fault-tolerant hybrid-parallel training at scale with reliable and efficient in-memory checkpointing","author":"Wang","year":"2023"},{"key":"ref16","article-title":"Optimize checkpoint performance for large models - Azure machine learning","year":"2025"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476205"},{"key":"ref18","article-title":"Megatron-LM: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3784828.3784830"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD58817.2023.00031"},{"key":"ref22","article-title":"Transom: An efficient fault-tolerant system for training LLMs","author":"Wu","year":"2023"},{"key":"ref23","first-page":"745","article-title":"MegaScale: Scaling model training to more than 10,000 GPUs","volume-title":"Proc. 21st USENIX Symp. Networked Syst. Des. Implementation","author":"Jiang","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613152"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3492321.3519584"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/PDCAT.2009.78"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2019.00099"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/HiPC56025.2022.00043"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3588195.3592987"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1016\/j.softx.2020.100561"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2010.18"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1145\/3018743.3018745"},{"key":"ref33","article-title":"An overview of gradient descent optimization algorithms","author":"Ruder","year":"2016"},{"key":"ref34","article-title":"Adam: A method for stochastic optimization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Kingma","year":"2015"},{"key":"ref35","article-title":"Mixed precision training","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Micikevicius","year":"2018"},{"key":"ref36","article-title":"Asynccheckpointio\u2013 pytorch lightning","year":"2024"},{"key":"ref37","article-title":"Source code repository","year":"2024"},{"key":"ref38","article-title":"DataStates: Lazy asynchronous checkpointing for large language models","year":"2024"},{"key":"ref39","article-title":"Polaris","year":"2025"},{"key":"ref40","first-page":"380","article-title":"Lustre: Building a file system for 1000-node clusters","volume-title":"Proc. Linux Symp.","author":"Schwan","year":"2003"},{"key":"ref41","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2022.3181799"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/71\/11609653\/11523632-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/71\/11609653\/11523632.pdf?arnumber=11523632","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T21:49:40Z","timestamp":1784238580000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11523632\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":42,"journal-issue":{"issue":"9"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2026.3694405","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"value":"1045-9219","type":"print"},{"value":"1558-2183","type":"electronic"},{"value":"2161-9883","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9]]}}}