{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T15:27:08Z","timestamp":1778081228117,"version":"3.51.4"},"reference-count":46,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2022,2,1]],"date-time":"2022-02-01T00:00:00Z","timestamp":1643673600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,2,1]],"date-time":"2022-02-01T00:00:00Z","timestamp":1643673600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,2,1]],"date-time":"2022-02-01T00:00:00Z","timestamp":1643673600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["CNS-1751075"],"award-info":[{"award-number":["CNS-1751075"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["CNS-1909306"],"award-info":[{"award-number":["CNS-1909306"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["CCF-1850029"],"award-info":[{"award-number":["CCF-1850029"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100004316","name":"2018 IBM Faculty Research Award","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004316","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100005144","name":"Qualcomm Innovation Fellowship","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100005144","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE\/ACM Trans. Networking"],"published-print":{"date-parts":[[2022,2]]},"DOI":"10.1109\/tnet.2021.3112082","type":"journal-article","created":{"date-parts":[[2021,11,10]],"date-time":"2021-11-10T21:57:12Z","timestamp":1636581432000},"page":"215-228","source":"Crossref","is-referenced-by-count":6,"title":["Machine Learning on Volatile Instances: Convergence, Runtime, and Cost Tradeoffs"],"prefix":"10.1109","volume":"30","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0751-2773","authenticated-orcid":false,"given":"Xiaoxi","family":"Zhang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7075-9333","authenticated-orcid":false,"given":"Jianyu","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Li-Feng","family":"Lee","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tom","family":"Yang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Akansha","family":"Kalra","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6372-9697","authenticated-orcid":false,"given":"Gauri","family":"Joshi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0785-9291","authenticated-orcid":false,"given":"Carlee","family":"Joe-Wong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155448"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TNET.2021.3112082"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177729586"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-7908-2604-3_16"},{"key":"ref5","first-page":"1223","article-title":"Large scale distributed deep networks","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst. (NIPS)","volume":"1","author":"Dean"},{"key":"ref6","volume-title":"Amazon EC2 Spot Instances","year":"2019"},{"key":"ref7","volume-title":"Preemptible Virtual Machines","year":"2019"},{"key":"ref8","volume-title":"Announcing Low-Priority VMS on Scale Sets Now in Public Preview","year":"2018"},{"key":"ref9","volume-title":"Spot Price Overrides","year":"2019"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.96"},{"key":"ref11","article-title":"Characterizing curtailed and uneconomic renewable power in the mid-continent independent system operator","volume-title":"arXiv:1702.05403","author":"Chien","year":"2016"},{"key":"ref12","article-title":"Federated learning: Strategies for improving communication efficiency","volume-title":"arXiv:1610.05492","author":"Kone\u010dn\u00fd","year":"2016"},{"key":"ref13","article-title":"eSGD: Communication efficient distributed deep learning on the edge","volume-title":"Proc. USENIX Workshop Hot Topics Edge Comput. (HotEdge)","author":"Tao"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155372"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/2785956.2787473"},{"key":"ref16","volume-title":"The CIFAR-10 Dataset","author":"Krizhevsky"},{"key":"ref17","article-title":"How not to bid the cloud","volume-title":"Proc. USENIX Conf. Hot Topics Cloud Comput. (HotCloud)","author":"Sharma"},{"key":"ref18","volume-title":"Preemptible VM Instances","year":"2019"},{"key":"ref19","first-page":"165","article-title":"Optimal distributed online prediction using mini-batches","volume":"13","author":"Dekel","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT44484.2020.9174216"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-10925-7_24"},{"key":"ref22","article-title":"Communication-efficient learning of deep networks from decentralized data","volume-title":"arXiv:1602.05629","author":"McMahan","year":"2016"},{"key":"ref23","first-page":"75","article-title":"Slow and stale gradients can win the race: Error-runtime trade-offs in distributed SGD","volume-title":"Proc. AISTATS","author":"Dutta"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1137\/16M1080173"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3310336"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CLOUD.2012.59"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3064176.3064182"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CLOUD.2017.21"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/2987550.2987576"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/2020408.2020410"},{"key":"ref31","article-title":"Adaptive communication strategies to achieve the best error-runtime trade-off in local-update SGD","volume-title":"Proc. SysML Conf.","author":"Wang"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.14778\/2732967.2732973"},{"key":"ref33","article-title":"Revisiting distributed synchronous SGD","volume-title":"Proc. ICLR Workshop Track","author":"Chen"},{"key":"ref34","first-page":"7174","article-title":"On the computation and communication complexity of parallel SGD with dynamic batch sizes for stochastic non-convex optimization","volume-title":"Proc. ICML","volume":"97","author":"Yu"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/2408776.2408794"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-26050-7_257-1"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/0-387-26336-5_540"},{"key":"ref38","volume-title":"How Spot Instances Work"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2019.2927558"},{"key":"ref40","article-title":"Deep residual learning for image recognition","volume-title":"arXiv:1512.03385","author":"He","year":"2015"},{"key":"ref41","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. NIPS","author":"Krizhevsky"},{"key":"ref42","first-page":"561","article-title":"Ray: A distributed framework for emerging ai applications","volume-title":"Proc. USENIX OSDI","author":"Moritz"},{"key":"ref43","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","volume-title":"Proc. USENIX OSDI","author":"Abadi"},{"key":"ref44","article-title":"Accurate, large minibatch SGD: Training ImageNet in 1 hour","volume-title":"arXiv:1706.02677","author":"Goyal","year":"2018"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46128-1_50"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1972.10482404"}],"container-title":["IEEE\/ACM Transactions on Networking"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/90\/9714545\/09610614.pdf?arnumber=9610614","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,12]],"date-time":"2024-01-12T02:11:01Z","timestamp":1705025461000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9610614\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2]]},"references-count":46,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tnet.2021.3112082","relation":{},"ISSN":["1063-6692","1558-2566"],"issn-type":[{"value":"1063-6692","type":"print"},{"value":"1558-2566","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,2]]}}}