{"id":"https://openalex.org/W4360831773","doi":"https://doi.org/10.1109/hpca56546.2023.10071024","title":"OptimStore: In-Storage Optimization of Large Scale DNNs with On-Die Processing","display_name":"OptimStore: In-Storage Optimization of Large Scale DNNs with On-Die Processing","publication_year":2023,"publication_date":"2023-02-01","ids":{"openalex":"https://openalex.org/W4360831773","doi":"https://doi.org/10.1109/hpca56546.2023.10071024"},"language":"en","primary_location":{"id":"doi:10.1109/hpca56546.2023.10071024","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10071024","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5066364128","display_name":"Junkyum Kim","orcid":"https://orcid.org/0009-0003-2519-2379"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Junkyum Kim","raw_affiliation_strings":["Samsung Electronics,SAIT","School of Electrical Engineering, KAIST","SAIT, Samsung Electronics"],"affiliations":[{"raw_affiliation_string":"Samsung Electronics,SAIT","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"SAIT, Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047372993","display_name":"Myeonggu Kang","orcid":"https://orcid.org/0000-0003-3557-8526"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Myeonggu Kang","raw_affiliation_strings":["KAIST,School of Electrical Engineering","School of Electrical Engineering, KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036617506","display_name":"Yunki Han","orcid":"https://orcid.org/0000-0003-0432-9324"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yunki Han","raw_affiliation_strings":["KAIST,School of Electrical Engineering","School of Electrical Engineering, KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089909294","display_name":"Yang-Gon Kim","orcid":"https://orcid.org/0009-0009-6165-6234"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]},{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yang-Gon Kim","raw_affiliation_strings":["KAIST,School of Electrical Engineering","System LSI, Samsung Electronics","School of Electrical Engineering, KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"System LSI, Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052390471","display_name":"Lee\u2010Sup Kim","orcid":"https://orcid.org/0000-0001-9585-4591"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Lee-Sup Kim","raw_affiliation_strings":["KAIST,School of Electrical Engineering","School of Electrical Engineering, KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5066364128"],"corresponding_institution_ids":["https://openalex.org/I157485424","https://openalex.org/I2250650973"],"apc_list":null,"apc_paid":null,"fwci":4.2989,"has_fulltext":false,"cited_by_count":21,"citation_normalized_percentile":{"value":0.94605882,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"611","last_page":"623"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9936000108718872,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9914000034332275,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.8363285064697266},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.812938392162323},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6741386651992798},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5246337056159973},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.5147009491920471},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.43771788477897644},{"id":"https://openalex.org/keywords/flash","display_name":"Flash (photography)","score":0.4229735732078552},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3280234634876251},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.3208213150501251},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3018415570259094},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2714451551437378},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.25457286834716797},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1005696952342987}],"concepts":[{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.8363285064697266},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.812938392162323},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6741386651992798},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5246337056159973},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.5147009491920471},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.43771788477897644},{"id":"https://openalex.org/C2777526259","wikidata":"https://www.wikidata.org/wiki/Q221836","display_name":"Flash (photography)","level":2,"score":0.4229735732078552},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3280234634876251},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3208213150501251},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3018415570259094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2714451551437378},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.25457286834716797},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1005696952342987},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca56546.2023.10071024","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10071024","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.8799999952316284}],"awards":[],"funders":[{"id":"https://openalex.org/F4320326288","display_name":"K2","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":78,"referenced_works":["https://openalex.org/W1474183241","https://openalex.org/W1522301498","https://openalex.org/W1608061514","https://openalex.org/W1686810756","https://openalex.org/W2004376136","https://openalex.org/W2097117768","https://openalex.org/W2146502635","https://openalex.org/W2170382128","https://openalex.org/W2177227417","https://openalex.org/W2219142176","https://openalex.org/W2249612659","https://openalex.org/W2265957089","https://openalex.org/W2558722810","https://openalex.org/W2763421725","https://openalex.org/W2766040387","https://openalex.org/W2767352027","https://openalex.org/W2785961331","https://openalex.org/W2884711234","https://openalex.org/W2899332646","https://openalex.org/W2901739041","https://openalex.org/W2903754802","https://openalex.org/W2920916186","https://openalex.org/W2949251802","https://openalex.org/W2954618783","https://openalex.org/W2962739339","https://openalex.org/W2977083675","https://openalex.org/W2979937837","https://openalex.org/W2985022115","https://openalex.org/W2989484363","https://openalex.org/W3010830594","https://openalex.org/W3012479151","https://openalex.org/W3012514909","https://openalex.org/W3019791076","https://openalex.org/W3042638450","https://openalex.org/W3080573571","https://openalex.org/W3105115497","https://openalex.org/W3112282291","https://openalex.org/W3129831491","https://openalex.org/W3133969675","https://openalex.org/W3138176046","https://openalex.org/W3138246627","https://openalex.org/W3155143326","https://openalex.org/W3155243801","https://openalex.org/W3155475857","https://openalex.org/W3156643189","https://openalex.org/W3157864729","https://openalex.org/W3158634533","https://openalex.org/W4214559301","https://openalex.org/W4226064548","https://openalex.org/W4246193833","https://openalex.org/W4251178606","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4293508727","https://openalex.org/W4311457721","https://openalex.org/W4385245566","https://openalex.org/W6628814951","https://openalex.org/W6631190155","https://openalex.org/W6637373629","https://openalex.org/W6677759287","https://openalex.org/W6681435938","https://openalex.org/W6685733532","https://openalex.org/W6688776908","https://openalex.org/W6691351710","https://openalex.org/W6737818604","https://openalex.org/W6745245109","https://openalex.org/W6745776437","https://openalex.org/W6748509488","https://openalex.org/W6748522178","https://openalex.org/W6753209298","https://openalex.org/W6756104738","https://openalex.org/W6765119170","https://openalex.org/W6769627184","https://openalex.org/W6778883912","https://openalex.org/W6790981935","https://openalex.org/W6791832985","https://openalex.org/W6794260887","https://openalex.org/W6810510860"],"related_works":["https://openalex.org/W1967627035","https://openalex.org/W2540591044","https://openalex.org/W4249402901","https://openalex.org/W1509211761","https://openalex.org/W2391299576","https://openalex.org/W2319467001","https://openalex.org/W1869243490","https://openalex.org/W2133693067","https://openalex.org/W1531488649","https://openalex.org/W3029399624"],"abstract_inverted_index":{"Training":[0],"deep":[1],"neural":[2],"network":[3],"(DNN)":[4],"models":[5,59,75,88,144],"is":[6,60,67,99],"a":[7,120,179,183],"resource-intensive,":[8],"iterative":[9],"process.":[10,113],"For":[11],"this":[12,115],"reason,":[13],"nowadays,":[14],"complex":[15],"optimizers":[16],"like":[17],"Adam":[18],"are":[19],"widely":[20],"adopted":[21],"as":[22,54],"it":[23,66],"increases":[24],"the":[25,39,49,55,73,94,97,103,110,138,150,155,163,189],"speed":[26],"and":[27,37,170,182],"efficiency":[28,187],"of":[29,45,57,141,160],"training.":[30],"These":[31],"optimizers,":[32],"however,":[33],"employ":[34],"additional":[35],"variables":[36],"raise":[38],"memory":[40,50],"demand":[41],"2\u00d7":[42],"to":[43,62,70,86,89,101],"3\u00d7":[44],"model":[46,147],"parameters,":[47],"worsening":[48],"capacity":[51],"bottleneck.":[52],"Moreover,":[53],"size":[56],"DNN":[58],"projected":[61],"grow":[63],"even":[64],"further,":[65],"not":[68],"practical":[69],"assume":[71],"that":[72],"future":[74],"will":[76],"fit":[77],"in":[78,149,188],"accelerator":[79],"memory.":[80],"This":[81],"has":[82],"triggered":[83],"various":[84],"efforts":[85],"offload":[87],"flash-based":[90],"storage.":[91],"However,":[92],"when":[93],"model,":[95],"especially":[96],"optimizer,":[98],"offloaded":[100],"flash,":[102],"limited":[104],"I/O":[105],"bandwidth":[106],"severely":[107],"slows":[108],"down":[109],"overall":[111],"training":[112,139],"To":[114],"end,":[116],"we":[117],"present":[118],"OptimStore,":[119],"solid-state":[121],"drive":[122],"(SSD)":[123],"system":[124],"with":[125],"on-die":[126],"processing":[127,146],"(ODP)":[128],"architectures":[129],"for":[130],"gradient":[131],"descent-based":[132],"machine":[133],"learning":[134],"models.":[135],"OptimStore":[136,161,175],"accelerates":[137],"process":[140],"such":[142],"large-scale":[143],"by":[145],"optimization":[148],"storage":[151],"device,":[152],"specifically":[153],"inside":[154],"flash":[156,172],"dies.":[157],"ODP":[158],"capability":[159],"eliminates":[162],"heavy":[164],"data":[165],"movement":[166],"over":[167,193],"external":[168],"interconnect":[169],"internal":[171],"channels.":[173],"Overall,":[174],"achieves,":[176],"on":[177],"average,":[178],"2.8\u00d7":[180],"speedup":[181],"3.6\u00d7":[184],"improved":[185],"energy":[186],"weight":[190],"update":[191],"stage":[192],"baseline":[194],"SSD":[195],"offloading.":[196]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":7},{"year":2023,"cited_by_count":2}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}