{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T04:13:21Z","timestamp":1782965601571,"version":"3.54.5"},"reference-count":46,"publisher":"IEEE","license":[{"start":{"date-parts":[[2020,7,1]],"date-time":"2020-07-01T00:00:00Z","timestamp":1593561600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2020,7,1]],"date-time":"2020-07-01T00:00:00Z","timestamp":1593561600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2020,7,1]],"date-time":"2020-07-01T00:00:00Z","timestamp":1593561600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,7]]},"DOI":"10.1109\/infocom41043.2020.9155446","type":"proceedings-article","created":{"date-parts":[[2020,8,4]],"date-time":"2020-08-04T22:29:35Z","timestamp":1596580175000},"page":"626-635","source":"Crossref","is-referenced-by-count":59,"title":["Preemptive All-reduce Scheduling for Expediting Distributed DNN Training"],"prefix":"10.1109","author":[{"given":"Yixin","family":"Bao","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanghua","family":"Peng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yangrui","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chuan","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1045"},{"key":"ref38","article-title":"3LC: Lightweight and Effective Traffic Compression for Distributed Machine Learning","author":"hyeontaek lim","year":"2019","journal-title":"Proc of SysML"},{"key":"ref33","article-title":"WMT 2017","year":"2017"},{"key":"ref32","article-title":"Sequence to Sequence Learning with Neural Networks","author":"sutskever","year":"2014","journal-title":"Proc of NIPS"},{"key":"ref31","article-title":"ImageNet Dataset","year":"2017"},{"key":"ref30","article-title":"Run Deep Learning with PaddlePaddle on Kubernetes","year":"2019"},{"key":"ref37","article-title":"Terngrad: Ternary Gradients to Reduce Communication in Distributed Deep Learning","author":"wen","year":"2017","journal-title":"Proc of NIPS"},{"key":"ref36","article-title":"AWS EC2 Instance","year":"2019"},{"key":"ref35","article-title":"MXNet sockeye","year":"2019"},{"key":"ref34","article-title":"MXNet Image Classification","year":"2019"},{"key":"ref10","article-title":"MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems","author":"chen","year":"2016","journal-title":"Proc NIPS Workshop Mach Learn Syst"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3190508.3190517"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359642"},{"key":"ref12","article-title":"Priority-Based Parameter Propagation for Distributed DNN Training","author":"jayarajan","year":"2019","journal-title":"Proc of SysML"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33015289"},{"key":"ref14","article-title":"PaddlePaddle","year":"2019"},{"key":"ref15","article-title":"Horovod: Fast and Easy Distributed Deep Learning in TensorFlow","author":"sergeev","year":"2018"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-7908-2604-3_16"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2008.09.002"},{"key":"ref18","author":"hintjens","year":"2013","journal-title":"Zeromq Messaging for Many Applications"},{"key":"ref19","article-title":"Very Deep Convolutional Networks for LargeScale Image Recognition","author":"karen simonyan","year":"2015","journal-title":"Proc of ICLR"},{"key":"ref28","article-title":"NVIDIA Collective Communications Library (NCCL)","year":"2019"},{"key":"ref4","article-title":"QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding","author":"alistarh","year":"2017","journal-title":"Proc of NIPS"},{"key":"ref27","article-title":"On the Importance of Initialization and Momentum in Deep Learning","author":"sutskever","year":"2013","journal-title":"Proc of ICML"},{"key":"ref3","article-title":"Poseidon: An Efficient Communication Architecture for Distributed Deep Learning on GPU Clusters","author":"zhang","year":"2017","journal-title":"Proc of USENIX ATC"},{"key":"ref6","article-title":"Mixed Precision Training","author":"micikevicius","year":"2018","journal-title":"Proc of ICLR"},{"key":"ref29","article-title":"Kubernetes","year":"2019"},{"key":"ref5","article-title":"TicTac: Accelerating Distributed Deep Learning with Communication Scheduling","author":"hashemi","year":"2019","journal-title":"Proc of SysML"},{"key":"ref8","article-title":"TensorFlow: A System for Large-Scale Machine Learning","author":"abadi","year":"2016","journal-title":"Proc of USENIX OSDI"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1023\/B:IJPP.0000029272.69895.c1"},{"key":"ref2","article-title":"Large Scale Distributed Deep Networks","author":"dean","year":"2012","journal-title":"Proc of NIPS"},{"key":"ref9","article-title":"Automatic Differentiation in PyTorch","author":"paszke","year":"2017","journal-title":"Proc NIPS Autodiff Workshop"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737367"},{"key":"ref46","article-title":"Optimizing Network Performance in Distributed Machine Learning","author":"mai","year":"2015","journal-title":"Proc of USENIX HotCloud"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737587"},{"key":"ref22","article-title":"Device Placement Optimization with Reinforcement Learning","author":"mirhoseini","year":"2017","journal-title":"Proc of ICML"},{"key":"ref21","article-title":"Attention is All You Need","author":"vaswani","year":"2017","journal-title":"Proc of NIPS"},{"key":"ref42","article-title":"DL2: A Deep Learning-driven Scheduler for Deep Learning Clusters","author":"peng","year":"2019"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2015.7218460"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2018.8486422"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.20.11.1442"},{"key":"ref44","article-title":"More Effective Distributed ML via a Stale Synchronous Parallel Parameter Server","author":"ho","year":"2013","journal-title":"Proc of NIPS"},{"key":"ref26","author":"bernhard","year":"2018","journal-title":"Combinatorial Optimization Theory and Algorithms"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737460"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1137\/0315059"}],"event":{"name":"IEEE INFOCOM 2020 - IEEE Conference on Computer Communications","location":"Toronto, ON, Canada","start":{"date-parts":[[2020,7,6]]},"end":{"date-parts":[[2020,7,9]]}},"container-title":["IEEE INFOCOM 2020 - IEEE Conference on Computer Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9146503\/9155217\/09155446.pdf?arnumber=9155446","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,6,28]],"date-time":"2022-06-28T21:55:40Z","timestamp":1656453340000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9155446\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,7]]},"references-count":46,"URL":"https:\/\/doi.org\/10.1109\/infocom41043.2020.9155446","relation":{},"subject":[],"published":{"date-parts":[[2020,7]]}}}