{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T01:13:52Z","timestamp":1740100432449,"version":"3.37.3"},"reference-count":23,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,9,5]],"date-time":"2021-09-05T00:00:00Z","timestamp":1630800000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,9,5]],"date-time":"2021-09-05T00:00:00Z","timestamp":1630800000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,9,5]],"date-time":"2021-09-05T00:00:00Z","timestamp":1630800000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key R&D Program of China","doi-asserted-by":"publisher","award":["2018YFB0204300"],"award-info":[{"award-number":["2018YFB0204300"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,9,5]]},"DOI":"10.1109\/iscc53001.2021.9631443","type":"proceedings-article","created":{"date-parts":[[2021,12,15]],"date-time":"2021-12-15T20:46:50Z","timestamp":1639601210000},"page":"1-7","source":"Crossref","is-referenced-by-count":0,"title":["FastHorovod: Expediting Parallel Message-Passing Schedule for Distributed DNN Training"],"prefix":"10.1109","author":[{"given":"Yanghai","family":"Wang","sequence":"first","affiliation":[]},{"given":"Dezun","family":"Dong","sequence":"additional","affiliation":[]},{"given":"Yemao","family":"Xu","sequence":"additional","affiliation":[]},{"given":"Shuo","family":"Ouyang","sequence":"additional","affiliation":[]},{"given":"Xiangke","family":"Liao","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref10","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"sergeev","year":"2018","journal-title":"ArXiv Preprint"},{"key":"ref11","first-page":"489","article-title":"mtcp: a highly scalable user-level TCP stack for multicore systems","author":"jeong","year":"2014","journal-title":"7th USENIX Symposium on Networked System Design and Implementation (NSDI)"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid.2015.131"},{"key":"ref13","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Advances in Neural Information Processing Systems (NIPS)"},{"key":"ref14","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015","journal-title":"International Conference on Learning Representations (ICLR)"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3312570"},{"key":"ref16","article-title":"Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"chen","year":"2015","journal-title":"ar Xiv preprint"},{"key":"ref17","first-page":"8026","article-title":"Pytorch: An imperative style, high-performance deep learning library","author":"paszke","year":"2019","journal-title":"Advances in Neural Information Processing Systems (NIPS)"},{"journal-title":"Horovod","year":"0","key":"ref18"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3417607"},{"key":"ref4","article-title":"Accurate, large minibatch sgd: Training imagenet in 1 hour","author":"goyal","year":"2017","journal-title":"ArXiv Preprint"},{"key":"ref3","article-title":"Pipedream: Fast and efficient pipeline parallel dnn training","author":"harlap","year":"2018","journal-title":"ArXiv Preprint"},{"key":"ref6","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes","author":"jia","year":"2018","journal-title":"ArXiv Preprint"},{"key":"ref5","first-page":"3252","article-title":"Error feedback fixes signsgd and other gradient compression schemes","author":"karimireddy","year":"2019","journal-title":"Proc of the International Conference on Machine Learning (ICML)"},{"key":"ref8","article-title":"Nccl 2.0","author":"jeaugey","year":"2017","journal-title":"GTC"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3302424.3303975"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2020.11.005"},{"key":"ref1","first-page":"279","article-title":"Accel-erating distributed reinforcement learning with in-switch computing","author":"li","year":"2019","journal-title":"ACM\/IEEE 46th Annual International Symposium on Computer Architecture (ISCA)"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33518-1_5"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CLUSTER.2018.00014"},{"journal-title":"Github Repository","year":"0","key":"ref22"},{"journal-title":"Paracloud","year":"0","key":"ref21"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"}],"event":{"name":"2021 IEEE Symposium on Computers and Communications (ISCC)","start":{"date-parts":[[2021,9,5]]},"location":"Athens, Greece","end":{"date-parts":[[2021,9,8]]}},"container-title":["2021 IEEE Symposium on Computers and Communications (ISCC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9631377\/9631248\/09631443.pdf?arnumber=9631443","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T16:54:24Z","timestamp":1652201664000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9631443\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,5]]},"references-count":23,"URL":"https:\/\/doi.org\/10.1109\/iscc53001.2021.9631443","relation":{},"subject":[],"published":{"date-parts":[[2021,9,5]]}}}