{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:18:24Z","timestamp":1787012304291,"version":"build-2736575974"},"reference-count":55,"publisher":"IEEE","license":[{"start":{"date-parts":[[2019,11,1]],"date-time":"2019-11-01T00:00:00Z","timestamp":1572566400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2019,11,1]],"date-time":"2019-11-01T00:00:00Z","timestamp":1572566400000},"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":[[2019,11]]},"DOI":"10.1109\/iiswc47752.2019.9042047","type":"proceedings-article","created":{"date-parts":[[2020,3,20]],"date-time":"2020-03-20T04:55:14Z","timestamp":1584680114000},"page":"189-202","source":"Crossref","is-referenced-by-count":41,"title":["Characterizing Deep Learning Training Workloads on Alibaba-PAI"],"prefix":"10.1109","author":[{"given":"Mengdi","family":"Wang","sequence":"first","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Meng","sequence":"additional","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guoping","family":"Long","sequence":"additional","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chuan","family":"Wu","sequence":"additional","affiliation":[{"name":"The University of Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Yang","sequence":"additional","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Lin","sequence":"additional","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yangqing","family":"Jia","sequence":"additional","affiliation":[{"name":"Alibaba Group"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","author":"qi","year":"2016","journal-title":"Paleo A performance model for deep neural networks"},{"key":"ref38","first-page":"3","article-title":"Evaluating on-node gpu interconnects for deep learning workloads","author":"tallent","year":"0","journal-title":"International Workshop on Performance Modeling Benchmarking and Simulation of High Performance Computer Systems"},{"key":"ref33","first-page":"583","article-title":"Scaling distributed machine learning with the parameter server","author":"li","year":"0","journal-title":"11th USENIX Symposium on Operating Systems Design and Implementation (OSDI 14)"},{"key":"ref32","author":"shallue","year":"2018","journal-title":"Measuring the effects of data parallelism on neural network training"},{"key":"ref31","author":"mayer","year":"2019","journal-title":"Scalable deep learning on distributed infrastructures Challenges techniques and tools"},{"key":"ref30","year":"0"},{"key":"ref37","author":"hashemi","year":"2018","journal-title":"Tictac Accelerating distributed deep learning with communication scheduling"},{"key":"ref36","first-page":"181","article-title":"Poseidon: An efficient communication architecture for distributed deep learning on {GPU} clusters","author":"zhang","year":"0","journal-title":"2017 USENIX Annual Technical Conference ( USENIX ATC 17)"},{"key":"ref35","author":"goldsborough","year":"2016","journal-title":"A Tour of Tensorflow"},{"key":"ref34","year":"2018","journal-title":"NVIDIA Collective Comm Library"},{"key":"ref28","author":"chen","year":"2015","journal-title":"Mxnet A flexible and efficient machine learning library for heterogeneous distributed systems"},{"key":"ref27","author":"paszke","year":"2017","journal-title":"On Automatic Differentiation"},{"key":"ref29","article-title":"An introduction to computational networks and the computational network toolkit","author":"yu","year":"0","journal-title":"Microsoft Technical Report MSR-TR-2014-112 2014"},{"key":"ref2","first-page":"630","article-title":"Identity mappings in deep residual networks","author":"he","year":"0","journal-title":"European Conference on Computer Vision"},{"key":"ref1","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":"ref20","author":"park","year":"2018","journal-title":"Deep learning inference in facebook data centers Characterization performance optimizations and hardware implications"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2018.8573476"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/DASC\/PiCom\/DataCom\/CyberSciTec.2018.000-4"},{"key":"ref24","year":"2017","journal-title":"Xla - tensorflow compiled post in the google developers blog"},{"key":"ref23","year":"2017","journal-title":"Nvidia tesla v100 gpu architecture"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"ref25","first-page":"265","article-title":"Tensorflow: A system for large-scale machine learning","author":"abadi","year":"0","journal-title":"12th USENIX Symposium on Operating Systems Design and Implementation ( OSDI 16)"},{"key":"ref50","author":"gao","year":"2018","journal-title":"Data Motifs A Lens Towards Fully Understanding Big Data and AI Workloads"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.24251\/HICSS.2018.702"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/BigData.2018.8622396"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS.2009.4919648"},{"key":"ref53","first-page":"363","article-title":"Ernest: efficient performance prediction for large-scale advanced analytics","author":"venkataraman","year":"0","journal-title":"13th USENIX Symposium on Networked Systems Design and Implementation ( NSDI 16)"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/PDSW-DISCS.2018.00011"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3219890"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.312"},{"key":"ref40","first-page":"379","article-title":"R-fcn: Object detection via region-based fully convolutional networks","author":"dai","year":"2016","journal-title":"Advances in neural information processing systems"},{"key":"ref12","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1038\/nature16961","article-title":"Mastering the game of go with deep neural networks and tree search","volume":"529","author":"silver","year":"2016","journal-title":"Nature"},{"key":"ref13","author":"zoph","year":"2016","journal-title":"Neural architecture search with reinforcement learning"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CCBD.2016.029"},{"key":"ref15","author":"gu","year":"2017","journal-title":"DeepProf Performance Analysis for Deep Learning Applications via Mining GPU Execution Patterns"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2016.7581275"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2018.8573483"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2018.8573475"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2018.00059"},{"key":"ref4","first-page":"568","article-title":"Two-stream convolutional networks for action recognition in videos","author":"simonyan","year":"2014","journal-title":"Advances in neural information processing systems"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.223"},{"key":"ref6","author":"devlin","year":"2018","journal-title":"BERT Pre-training of deep bidirectional transformers for language understanding"},{"key":"ref5","author":"bahdanau","year":"2014","journal-title":"Neural machine translation by jointly learning to align and translate"},{"key":"ref8","first-page":"577","article-title":"Attention-based models for speech recognition","author":"chorowski","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3132747.3132772"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3219869"},{"key":"ref46","author":"micikevicius","year":"2017","journal-title":"Mixed Precision Training"},{"key":"ref45","author":"ruder","year":"2016","journal-title":"An overview of gradient descent optimization algorithms"},{"key":"ref48","article-title":"Multi-tenant gpu clusters for deep learning workloads: Analysis and implications","author":"jeon","year":"2018","journal-title":"Technical report MSR-T R-2018"},{"key":"ref47","author":"long","year":"2018","journal-title":"Fusionstitching Deep fusion and code generation for tensorflow computations on gpus"},{"key":"ref42","author":"kim","year":"2017","journal-title":"Dynamic layer normalization for adaptive neural acoustic modeling in speech recognition"},{"key":"ref41","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/2507157.2507209"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/2959100.2959190"}],"event":{"name":"2019 IEEE International Symposium on Workload Characterization (IISWC)","location":"Orlando, FL, USA","start":{"date-parts":[[2019,11,3]]},"end":{"date-parts":[[2019,11,5]]}},"container-title":["2019 IEEE International Symposium on Workload Characterization (IISWC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9027808\/9041927\/09042047.pdf?arnumber=9042047","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T18:26:55Z","timestamp":1755800815000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9042047\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,11]]},"references-count":55,"URL":"https:\/\/doi.org\/10.1109\/iiswc47752.2019.9042047","relation":{},"subject":[],"published":{"date-parts":[[2019,11]]}}}