{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T16:05:31Z","timestamp":1780589131032,"version":"3.54.1"},"reference-count":36,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,5,2]],"date-time":"2022-05-02T00:00:00Z","timestamp":1651449600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,5,2]],"date-time":"2022-05-02T00:00:00Z","timestamp":1651449600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100006190","name":"Research and Development","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006190","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,5,2]]},"DOI":"10.1109\/infocom48880.2022.9796820","type":"proceedings-article","created":{"date-parts":[[2022,6,20]],"date-time":"2022-06-20T21:18:49Z","timestamp":1655759929000},"page":"350-359","source":"Crossref","is-referenced-by-count":12,"title":["Mercury: A Simple Transport Layer Scheduler to Accelerate Distributed DNN Training"],"prefix":"10.1109","author":[{"given":"Qingyang","family":"Duan","sequence":"first","affiliation":[{"name":"School of Information Science and Technology, Fudan University,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zeqin","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Fudan University,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuedong","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Fudan University,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaoteng","family":"Liu","sequence":"additional","affiliation":[{"name":"Huawei Technologies Co. Ltd.,Shenzhen,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Fudan University,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref33","article-title":"Motivating in-network aggregation for distributed deep neural network training","author":"luo","year":"2017","journal-title":"Workshop on Approximate Computing Across the System Stack"},{"key":"ref32","article-title":"Hogwild!: A lock-free approach to parallelizing stochastic gradient descent","author":"niu","year":"2011","journal-title":"arXiv preprint arXiv 1106 5730"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737587"},{"key":"ref30","first-page":"1223","article-title":"More effective distributed ML via a stale synchronous parallel parameter server","author":"ho","year":"2013","journal-title":"Advances in neural information processing systems"},{"key":"ref36","first-page":"741","article-title":"ATP: In-network aggregation for multi-tenant learning","author":"lao","year":"2021","journal-title":"USENIX NSDI"},{"key":"ref35","first-page":"785","article-title":"Scaling distributed machine learning with in-network aggregation","author":"sapio","year":"2021","journal-title":"USENIX NSDI"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3152434.3152461"},{"key":"ref10","first-page":"181","article-title":"Poseidon: An efficient communication architecture for distributed deep learning on GPU clusters","author":"zhang","year":"2017","journal-title":"USENIX ATC"},{"key":"ref11","article-title":"TicTac: Accelerating distributed deep learning with communication scheduling","author":"hashemi","year":"2019","journal-title":"Proc of MLSys 2019"},{"key":"ref12","article-title":"Priority-based parameter propagation for distributed DNN training","author":"jayarajan","year":"2019","journal-title":"Proc of MLSys 2019"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359642"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737367"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3052862"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM42981.2021.9488803"},{"key":"ref17","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2014","journal-title":"arXiv preprint arXiv 1409 1556"},{"key":"ref18","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"CoRR"},{"key":"ref19","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"sergeev","year":"2018","journal-title":"arXiv preprint arxiv 1802 05807"},{"key":"ref28","article-title":"ScaleCom: Scalable sparsified gradient compression for communication-efficient distributed training","volume":"33","author":"chen","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref4","first-page":"173","article-title":"Deep speech 2: End-to-end speech recognition in English and Mandarin","author":"amodei","year":"2016","journal-title":"International Conference on Machine Learning"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00220"},{"key":"ref3","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"devlin","year":"2019","journal-title":"Proc of ACL"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.011.2000530"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155269"},{"key":"ref5","first-page":"583","article-title":"Scaling distributed machine learning with the parameter server","author":"li","year":"2014","journal-title":"USENIX OSDI"},{"key":"ref8","first-page":"8026","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"paszke","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref7","article-title":"MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"chen","year":"2015","journal-title":"CoRR"},{"key":"ref2","first-page":"177","article-title":"Decoding with value networks for neural machine translation","author":"he","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref9","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","author":"abadi","year":"2016","journal-title":"USENIX OSDI"},{"key":"ref1","first-page":"4188","article-title":"Gradually updated neural networks for large-scale image recognition","author":"qiao","year":"2018","journal-title":"International Conference on Machine Learning"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3386367.3431307"},{"key":"ref21","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3062721"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155282"},{"key":"ref26","first-page":"1709","article-title":"QSGD: Communication-efficient SGD via gradient quantization and encoding","volume":"30","author":"alistarh","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref25","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","author":"wen","year":"2017","journal-title":"Advances in neural information processing systems"}],"event":{"name":"IEEE INFOCOM 2022 - IEEE Conference on Computer Communications","location":"London, United Kingdom","start":{"date-parts":[[2022,5,2]]},"end":{"date-parts":[[2022,5,5]]}},"container-title":["IEEE INFOCOM 2022 - IEEE Conference on Computer Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9796607\/9796652\/09796820.pdf?arnumber=9796820","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,11]],"date-time":"2022-07-11T20:01:19Z","timestamp":1657569679000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9796820\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,2]]},"references-count":36,"URL":"https:\/\/doi.org\/10.1109\/infocom48880.2022.9796820","relation":{},"subject":[],"published":{"date-parts":[[2022,5,2]]}}}