{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,10,30]],"date-time":"2024-10-30T11:29:03Z","timestamp":1730287743215,"version":"3.28.0"},"reference-count":24,"publisher":"IEEE","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,12]]},"DOI":"10.1109\/padsw.2018.8644531","type":"proceedings-article","created":{"date-parts":[[2019,2,21]],"date-time":"2019-02-21T23:23:38Z","timestamp":1550791418000},"page":"473-480","source":"Crossref","is-referenced-by-count":2,"title":["Optimizing Deep Learning Frameworks Incrementally to Get Linear Speedup: A Comparison Between IPoIB and RDMA Verbs"],"prefix":"10.1109","author":[{"given":"Chang","family":"Liu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianwen","family":"Wei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yi-Chao","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Minhua","family":"Wen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Simon","family":"See","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"James","family":"Lin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","first-page":"445","author":"spring","year":"2017","journal-title":"Scalable and Sustainable Deep Learning via Randomized Hasing"},{"key":"ref11","first-page":"103","author":"mitchell","year":"2013","journal-title":"Using one-sided rdma reads to build a fast cpu-efficient key-value store"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1023\/B:IJPP.0000029272.69895.c1"},{"key":"ref13","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"sergeev","year":"2018","journal-title":"arXiv preprint arxiv 1802 05807"},{"key":"ref14","article-title":"Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"chen","year":"2015","journal-title":"arXiv preprint arXiv 1512 00327"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-7908-2604-3_16"},{"key":"ref16","article-title":"O'Reilly Media, Inc","author":"hintjens","year":"2013","journal-title":"Zeromq Messaging for Many Applications"},{"key":"ref17","first-page":"265","article-title":"Tensorflow: a system for large-scale machine learning","volume":"16","author":"abadi","year":"2016","journal-title":"OSDI"},{"key":"ref18","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/HPCC-SmartCity-DSS.2017.30"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1005177"},{"key":"ref3","first-page":"160","author":"collobert","year":"2008","journal-title":"A Unified Architecture for Natural Language Processing Deep Neural Networks with Multitask Learning"},{"key":"ref6","first-page":"19","article-title":"Distributed trammg of deep neural networks: theoretical and practical limits of parallel scalability","author":"keuper","year":"2016","journal-title":"Proceedings of the Workshop on Machine Learning in High Performance Computing Environments IEEE Press"},{"key":"ref5","article-title":"Distributed deep learning using synchronous stochastic gradient descent","author":"das","year":"2016","journal-title":"arXiv preprint arX-iv 1602 06709"},{"key":"ref8","article-title":"Squeezenet: Alexnet -level accuracy with 50x fewer parameters and] 0.5 mb model size","author":"landola","year":"2016","journal-title":"arXiv preprint arXiv 1602 00377"},{"key":"ref7","first-page":"1737","author":"gupta","year":"2015","journal-title":"Deep Learning with Limited Numerical Precision"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.2205597"},{"key":"ref1","first-page":"2818","author":"szegedy","year":"2016","journal-title":"Rethinking the inception architecture for computer vision"},{"key":"ref9","first-page":"2592","author":"landola","year":"2016","journal-title":"Firecaffe near-linear acceleration of deep neural network training on compute clusters"},{"key":"ref20","article-title":"Rpc considered harmful: Fast distributed deep learning on rdma","author":"xue","year":"2018","journal-title":"ar Xiv preprint arXiv 1805 08430"},{"key":"ref22","first-page":"16","author":"liu","year":"2004","journal-title":"Design and implementation of mpich2 over infiniband with rdma support"},{"key":"ref21","first-page":"28","author":"yi","year":"2017","journal-title":"Towards zero copy dataflows using rdma"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3155284.3018769"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2012.65"}],"event":{"name":"2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS)","start":{"date-parts":[[2018,12,11]]},"location":"Singapore, Singapore","end":{"date-parts":[[2018,12,13]]}},"container-title":["2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8635632\/8644527\/08644531.pdf?arnumber=8644531","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,27]],"date-time":"2022-01-27T05:40:18Z","timestamp":1643262018000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/8644531\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,12]]},"references-count":24,"URL":"https:\/\/doi.org\/10.1109\/padsw.2018.8644531","relation":{},"subject":[],"published":{"date-parts":[[2018,12]]}}}