{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T22:44:41Z","timestamp":1781131481889,"version":"3.54.1"},"reference-count":58,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2021,8,1]],"date-time":"2021-08-01T00:00:00Z","timestamp":1627776000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,8,1]],"date-time":"2021-08-01T00:00:00Z","timestamp":1627776000000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,8,1]],"date-time":"2021-08-01T00:00:00Z","timestamp":1627776000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,8,1]],"date-time":"2021-08-01T00:00:00Z","timestamp":1627776000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["1719160"],"award-info":[{"award-number":["1719160"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["1725447"],"award-info":[{"award-number":["1725447"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["1730309"],"award-info":[{"award-number":["1730309"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100004358","name":"Samsung Semiconductor, Inc","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004358","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst."],"published-print":{"date-parts":[[2021,8]]},"DOI":"10.1109\/tcad.2020.3021336","type":"journal-article","created":{"date-parts":[[2020,9,2]],"date-time":"2020-09-02T20:27:58Z","timestamp":1599078478000},"page":"1586-1599","source":"Crossref","is-referenced-by-count":18,"title":["DLUX: A LUT-Based Near-Bank Accelerator for Data Center Deep Learning Training Workloads"],"prefix":"10.1109","volume":"40","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2663-4568","authenticated-orcid":false,"given":"Peng","family":"Gu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7285-6682","authenticated-orcid":false,"given":"Xinfeng","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuangchen","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dimin","family":"Niu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongzhong","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Krishna T.","family":"Malladi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2093-1788","authenticated-orcid":false,"given":"Yuan","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","first-page":"33","article-title":"CACTI-3DD: Architecture-level modeling for 3d die-stacked DRAM main memory","author":"chen","year":"2012","journal-title":"Proc Design Autom Test Eur Conf Exhibition (DATE)"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2015.2414456"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2857044"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1145\/3243176.3243188"},{"key":"ref31","author":"kurup","year":"2012","journal-title":"Logic Synthesis Using Synopsys"},{"key":"ref30","year":"2018","journal-title":"Berkeley Hardware Floating-Point Units"},{"key":"ref37","author":"chetlur","year":"2014","journal-title":"cuDNN Efficient Primitives for Deep Learning"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/2366231.2337202"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/MDT.2011.44"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3357526.3357532"},{"key":"ref28","year":"2013","journal-title":"High-bandwidth Memory (HBM) DRAM"},{"key":"ref27","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","author":"abadi","year":"2016","journal-title":"Proc of USENIX Symp on Operating Systems Design and Implementation (OSDI)"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/HOTCHIPS.2011.7477494"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2018.00059"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-9260(99)00006-1"},{"key":"ref22","author":"hannun","year":"2014","journal-title":"Deep speech Scaling up end-to-end speech recognition"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3038912.3052569"},{"key":"ref24","first-page":"379","article-title":"A neural attention model for sentence summarization","author":"rush","year":"2015","journal-title":"Proc Conf Empirical Methods Nat Language Process"},{"key":"ref23","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.577"},{"key":"ref25","first-page":"3294","article-title":"Skip-thought vectors","author":"kiros","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080215"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2016.7783720"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/3037697.3037702"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001178"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/3307650.3322237"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2017.55"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001140"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001177"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2014.58"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2858358"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2017.2752706"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS.2014.6844483"},{"key":"ref12","author":"schuiki","year":"2018","journal-title":"A Scalable Near-Memory Architecture for Training Deep Neural Networks on Large In-Memory Datasets"},{"key":"ref13","article-title":"Scaling deep learning on multiple in-memory processors","author":"xu","year":"2015","journal-title":"Proc 3rd Workshop Near-Data Process Conjunction MICRO-48"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3154484"},{"key":"ref15","author":"nicol","year":"2017","journal-title":"A Dataflow Processing Chip for Training Deep Neural Networks"},{"key":"ref16","year":"2013","journal-title":"Hybrid Memory Cube Specification 2 1"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2602221"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3123939.3124545"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA45697.2020.00071"},{"key":"ref4","author":"jia","year":"2018","journal-title":"Dissecting the NVIDIA Volta GPU Architecture via Microbenchmarking"},{"key":"ref3","author":"zhu","year":"2018","journal-title":"Tbd Benchmarking and analyzing deep neural network training"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2016.7581275"},{"key":"ref5","year":"2018","journal-title":"MLPerf"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/1498765.1498785"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2019.2898196"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080221"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2018.00059"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/40.592312"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3195970.3199847"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001138"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1145\/3123939.3123977"},{"key":"ref42","first-page":"10","article-title":"1.1 computing’s energy problem (and what we can do about it)","author":"horowitz","year":"2014","journal-title":"IEEE Int Solid-State Circuits Conf Dig Techn Papers (ISSCC)"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080244"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3295500.3356156"},{"key":"ref43","author":"simonyan","year":"2017","journal-title":"Very Deep Convolutional Networks for Large-scale Image Recognition"}],"container-title":["IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/43\/9488253\/9184818-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/43\/9488253\/09184818.pdf?arnumber=9184818","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T14:49:30Z","timestamp":1652194170000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9184818\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,8]]},"references-count":58,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tcad.2020.3021336","relation":{},"ISSN":["0278-0070","1937-4151"],"issn-type":[{"value":"0278-0070","type":"print"},{"value":"1937-4151","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,8]]}}}