{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T15:37:18Z","timestamp":1772725038135,"version":"3.50.1"},"reference-count":65,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,2,1]],"date-time":"2023-02-01T00:00:00Z","timestamp":1675209600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,2,1]],"date-time":"2023-02-01T00:00:00Z","timestamp":1675209600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100010238","name":"K2","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100010238","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,2]]},"DOI":"10.1109\/hpca56546.2023.10071024","type":"proceedings-article","created":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T17:42:55Z","timestamp":1679679775000},"page":"611-623","source":"Crossref","is-referenced-by-count":19,"title":["OptimStore: In-Storage Optimization of Large Scale DNNs with On-Die Processing"],"prefix":"10.1109","author":[{"given":"Junkyum","family":"Kim","sequence":"first","affiliation":[{"name":"Samsung Electronics,SAIT"}]},{"given":"Myeonggu","family":"Kang","sequence":"additional","affiliation":[{"name":"KAIST,School of Electrical Engineering"}]},{"given":"Yunki","family":"Han","sequence":"additional","affiliation":[{"name":"KAIST,School of Electrical Engineering"}]},{"given":"Yang-Gon","family":"Kim","sequence":"additional","affiliation":[{"name":"KAIST,School of Electrical Engineering"}]},{"given":"Lee-Sup","family":"Kim","sequence":"additional","affiliation":[{"name":"KAIST,School of Electrical Engineering"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1","article-title":"Leveraging near data processing for high-performance checkpoint\/restart","volume-title":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","author":"Agrawal"},{"key":"ref2","first-page":"387","article-title":"Flashneuron: Ssd-enabled large-batch training of very deep neural networks","volume-title":"19th USENIX Conference on File and Storage Technologies (FAST 21)","author":"Bae"},{"key":"ref3","article-title":"Language models are few-shot learners","author":"Brown","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/DATE.2012.6176524"},{"key":"ref5","article-title":"Near-data processing for differentiable machine learning models","author":"Choe","year":"2016"},{"key":"ref6","article-title":"Multi-stream write ssd","volume-title":"Flash Memory Summit","author":"Choi","year":"2016"},{"key":"ref7","article-title":"Kioxia technical brief"},{"key":"ref8","article-title":"Mixed precision training of convolutional neural networks using integer operations","author":"Das","year":"2018"},{"key":"ref9","article-title":"Rmsprop and equilibrated adaptive learning rates for non-convex optimization","volume":"abs\/1502.04390","author":"Dauphin","year":"2015","journal-title":"CoRR"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/2463676.2465295"},{"key":"ref11","first-page":"2121","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"Duchi","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/MCAS.2020.2988388"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2018.00045"},{"key":"ref14","first-page":"1","article-title":"Thrifty: Training with hyperdimensional computing across flash hierarchy","volume-title":"2020 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)","author":"Gupta"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378465"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378530"},{"key":"ref17","article-title":"Onfi 5.0 specification"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2019.2953646"},{"key":"ref19","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes","author":"Jia","year":"2018"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC.2016.7761588"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2019.8662493"},{"key":"ref22","article-title":"DRAMPower: Open-source DRAM Power Energy Estimation Tool","author":"Chandrasekar"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00030"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3409963.3410501"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3409963.3410501"},{"key":"ref26","first-page":"371","article-title":"Behemoth: A flash-centric training accelerator for extreme-scale dnns","volume-title":"19th USENIX Conference on File and Storage Technologies (FAST 21)","author":"Kim"},{"key":"ref27","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref28","first-page":"219","article-title":"Summarizer: trading communication with computing near storage","volume-title":"2017 50th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)","author":"Koo"},{"key":"ref29","first-page":"219","article-title":"Summarizer: trading communication with computing near storage","volume-title":"2017 50th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)","author":"Koo"},{"key":"ref30","article-title":"Mixed-precision training for nlp and speech recognition with openseq2seq","author":"Kuchaiev","year":"2018"},{"key":"ref31","article-title":"Hardware\/software co-programmable framework for computational ssds to accelerate deep learning service on large-scale graphs","author":"Kwon","year":"2022"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2018.00021"},{"key":"ref33","first-page":"225","article-title":"Glist: Towards instorage graph learning","volume-title":"2021 USENIX Annual Technical Conference (USENIX ATC 21)","author":"Li"},{"key":"ref34","first-page":"469","article-title":"Mcpat: An integrated power, area, and timing modeling framework for multicore and manycore architectures","volume-title":"Proceedings of the 42nd annual ieee\/acm international symposium on microarchitecture","author":"Li"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2019.00035"},{"key":"ref36","first-page":"395","article-title":"Cognitive {SSD}: A deep learning engine for {In-Storage} data retrieval","volume-title":"2019 USENIX Annual Technical Conference (USENIX ATC 19)","author":"Liang"},{"issue":"10","key":"ref37","first-page":"00","article-title":"Optimizing nand flash-based ssds via retention relaxation","volume":"11","author":"Liu","year":"2012","journal-title":"Target"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/MSST.2015.7208284"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3224432"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3352460.3358320"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507702"},{"key":"ref42","article-title":"Nand flash memory having internal ecc processing and mehtod of operation thereof","volume-title":"U.S. Patent","author":"Michael","year":"2016"},{"key":"ref43","article-title":"Mixed precision training","volume-title":"International Conference on Learning Representations","author":"Micikevicius"},{"key":"ref44","article-title":"DeepSpeed"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446719"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/iiswc55918.2022.00033"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378505"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N18-1202"},{"issue":"8","key":"ref49","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref50","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","author":"Raffel","year":"2019"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref52","article-title":"Zeroinfinity: Breaking the gpu memory wall for extreme scale deep learning","author":"Rajbhandari","year":"2021"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00057"},{"key":"ref54","article-title":"Firecuda530 pcie 4.0 ssd"},{"key":"ref55","first-page":"67","article-title":"Willow: A user-programmable ssd","volume-title":"11th USENIX Symposium on Operating Systems Design and Implementation (OSDI 14)","author":"Seshadri"},{"key":"ref56","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/MASCOTS.2019.00025"},{"key":"ref58","article-title":"Design compiler"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605337-29"},{"key":"ref60","first-page":"119","article-title":"Active flash: Towards {Energy-Efficient}, {In-Situ} data analytics on {Extreme-Scale} machines","volume-title":"11th USENIX Conference on File and Storage Technologies (FAST 13)","author":"Tiwari"},{"key":"ref61","article-title":"Reducing data movement costs using {Energy-Efficient}, active computation on {SSD}","volume-title":"2012 Workshop on Power-Aware Computing and Systems (HotPower 12)","author":"Tiwari"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref63","article-title":"Improving the accuracy of the fast inverse square root algorithm","author":"Walczyk","year":"2018"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446763"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA45697.2020.00063"}],"event":{"name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","location":"Montreal, QC, Canada","start":{"date-parts":[[2023,2,25]]},"end":{"date-parts":[[2023,3,1]]}},"container-title":["2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10070856\/10070923\/10071024.pdf?arnumber=10071024","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,13]],"date-time":"2024-02-13T13:18:49Z","timestamp":1707830329000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10071024\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,2]]},"references-count":65,"URL":"https:\/\/doi.org\/10.1109\/hpca56546.2023.10071024","relation":{},"subject":[],"published":{"date-parts":[[2023,2]]}}}