{"id":"https://openalex.org/W4293731257","doi":"https://doi.org/10.1109/mm.2022.3202350","title":"Accelerating Neural Network Inference With Processing-in-DRAM: From the Edge to the Cloud","display_name":"Accelerating Neural Network Inference With Processing-in-DRAM: From the Edge to the Cloud","publication_year":2022,"publication_date":"2022-08-29","ids":{"openalex":"https://openalex.org/W4293731257","doi":"https://doi.org/10.1109/mm.2022.3202350"},"language":"en","primary_location":{"id":"doi:10.1109/mm.2022.3202350","doi":"https://doi.org/10.1109/mm.2022.3202350","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2022.3202350","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-a","jufo-3","norway-2","poland-140","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2209.08938","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5091016237","display_name":"Geraldo Francisco de Oliveira Junior","orcid":"https://orcid.org/0000-0003-1557-4819","observed_orcids":["https://orcid.org/0000-0003-1557-4819"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Geraldo F. Oliveira","raw_affiliation_strings":["ETH Zürich, Zurich, Switzerland"],"raw_orcid":"https://orcid.org/0000-0003-1557-4819","affiliations":[{"raw_affiliation_string":"ETH Zürich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044416322","display_name":"Juan G\u00f3mez-Luna","orcid":"https://orcid.org/0000-0002-6514-1571","observed_orcids":["https://orcid.org/0000-0002-6514-1571"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Juan Gomez-Luna","raw_affiliation_strings":["ETH Zürich, Zurich, Switzerland"],"raw_orcid":"https://orcid.org/0000-0002-6514-1571","affiliations":[{"raw_affiliation_string":"ETH Zürich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036666743","display_name":"Saugata Ghose","orcid":"https://orcid.org/0000-0002-9138-0613","observed_orcids":["https://orcid.org/0000-0002-9138-0613"]},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Saugata Ghose","raw_affiliation_strings":["University of Illinois Urbana–Champaign, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-9138-0613","affiliations":[{"raw_affiliation_string":"University of Illinois Urbana–Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016220809","display_name":"Amirali Boroumand","orcid":"https://orcid.org/0000-0001-9748-845X","observed_orcids":["https://orcid.org/0000-0001-9748-845X"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amirali Boroumand","raw_affiliation_strings":["Google, Mountain View, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google, Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050695684","display_name":"Onur Mutlu","orcid":"https://orcid.org/0000-0002-0075-2312","observed_orcids":["https://orcid.org/0000-0002-0075-2312"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Onur Mutlu","raw_affiliation_strings":["ETH Zürich, Zurich, Switzerland"],"raw_orcid":"https://orcid.org/0000-0002-0075-2312","affiliations":[{"raw_affiliation_string":"ETH Zürich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2995,"currency":"USD","value_usd":2995},"apc_paid":null,"fwci":5.1309,"has_fulltext":true,"cited_by_count":41,"citation_normalized_percentile":{"value":0.96388309,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":88,"max":100},"biblio":{"volume":"42","issue":"6","first_page":"25","last_page":"38"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.7851999998092651,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.7851999998092651,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.16110000014305115,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10363","display_name":"Low-power high-performance VLSI design","score":0.009800000116229057,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/in-memory-computing","display_name":"in-memory computing","score":0.9660000205039978},{"id":"https://openalex.org/keywords/dram","display_name":"DRAM","score":0.7680000066757202},{"id":"https://openalex.org/keywords/neural-network-inference-acceleration","display_name":"neural network inference acceleration","score":0.8889999985694885},{"id":"https://openalex.org/keywords/upmem","display_name":"UPMEM","score":0.8119999766349792},{"id":"https://openalex.org/keywords/mensa","display_name":"Mensa","score":0.9340000152587891},{"id":"https://openalex.org/keywords/3d-stacking","display_name":"3D stacking","score":0.8790000081062317},{"id":"https://openalex.org/keywords/edge-devices","display_name":"edge devices","score":0.6110000014305115},{"id":"https://openalex.org/keywords/energy-efficiency","display_name":"energy efficiency","score":0.6970000267028809},{"id":"https://openalex.org/keywords/binary-neural-networks","display_name":"binary neural networks","score":0.8180000185966492},{"id":"https://openalex.org/keywords/matrix-vector-multiplication","display_name":"matrix-vector multiplication","score":0.8349999785423279}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8673372268676758},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.6287013292312622},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.559431791305542},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.533652126789093},{"id":"https://openalex.org/C2779602883","wikidata":"https://www.wikidata.org/wiki/Q15544750","display_name":"Memory architecture","level":2,"score":0.43833670020103455},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3892630934715271},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.33031782507896423},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.32907646894454956}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/mm.2022.3202350","doi":"https://doi.org/10.1109/mm.2022.3202350","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2022.3202350","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-a","jufo-3","norway-2","poland-140","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2209.08938","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.08938","pdf_url":"https://arxiv.org/pdf/2209.08938","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2209.08938","doi":"https://doi.org/10.48550/arxiv.2209.08938","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2209.08938","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2209.08938","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.08938","pdf_url":"https://arxiv.org/pdf/2209.08938","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320306087","display_name":"Semiconductor Research Corporation","ror":"https://ror.org/047z4n946"},{"id":"https://openalex.org/F4320316785","display_name":"VMware","ror":"https://ror.org/01mwmh458"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4293731257.pdf","grobid_xml":"https://content.openalex.org/works/W4293731257.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W755906292","https://openalex.org/W1981943579","https://openalex.org/W2086112773","https://openalex.org/W2170257519","https://openalex.org/W2332254524","https://openalex.org/W2559524942","https://openalex.org/W2604319603","https://openalex.org/W2605347906","https://openalex.org/W2766489088","https://openalex.org/W2789554134","https://openalex.org/W2904200161","https://openalex.org/W2946571646","https://openalex.org/W2980688670","https://openalex.org/W2982008795","https://openalex.org/W3009123798","https://openalex.org/W3016765354","https://openalex.org/W3111721177","https://openalex.org/W3115641744","https://openalex.org/W3161639821","https://openalex.org/W3201951111","https://openalex.org/W4214734582","https://openalex.org/W4249322926","https://openalex.org/W4285121610"],"related_works":["https://openalex.org/W2943178787","https://openalex.org/W3127930667","https://openalex.org/W2920081941","https://openalex.org/W4416303630","https://openalex.org/W6889815743","https://openalex.org/W3138893848","https://openalex.org/W4378801005","https://openalex.org/W3211168157","https://openalex.org/W4379115511","https://openalex.org/W3163905326"],"abstract_inverted_index":{"Neural":[0],"networks":[1],"(NNs)":[2],"are":[3],"growing":[4],"in":[5,49],"importance":[6],"and":[7,65,76,93,111,164,168,183],"complexity.":[8],"An":[9],"NN\u2019s":[10],"performance":[11,75,139],"(and":[12],"energy":[13,77,162],"efficiency)":[14],"can":[15],"be":[16],"bound":[17],"either":[18],"by":[19,166,193],"computation":[20,29],"or":[21,33],"memory":[22,35,69,151],"resources.":[23],"The":[24],"processing-in-memory":[25],"(PIM)":[26],"paradigm,":[27],"where":[28,51],"is":[30,37,61],"placed":[31],"near":[32],"within":[34],"arrays,":[36],"a":[38,97,103,141,154,187,211],"viable":[39],"solution":[40],"to":[41,56,62,121,216],"accelerate":[42],"memory-bound":[43,132],"NNs.":[44,198],"However,":[45],"PIM":[46,53,71,86,105,129,204],"architectures":[47,72],"vary":[48],"form,":[50],"different":[52,57],"approaches":[54],"lead":[55],"tradeoffs.":[58],"Our":[59,125],"goal":[60],"analyze,":[63],"discuss,":[64],"contrast":[66],"dynamic":[67],"random-access":[68],"(DRAM)-based":[70],"for":[73,108,153,177,195,206],"NN":[74,181,207],"efficiency.":[78],"To":[79],"do":[80],"so,":[81],"we":[82],"analyze":[83],"three":[84,196],"state-of-the-art":[85],"architectures:":[87],"1)":[88,134],"UPMEM,":[89],"which":[90,114],"integrates":[91],"processors":[92],"DRAM":[94,120],"arrays":[95],"into":[96],"single":[98],"2-D":[99],"chip,":[100],"2)":[101,159],"Mensa,":[102],"3-D-stacking-based":[104],"architecture":[106,205],"tailored":[107],"edge":[109,180],"devices,":[110],"3)":[112,184],"SIMDRAM,":[113],"uses":[115],"the":[116,138,148,171,202,217],"analog":[117],"principles":[118],"of":[119,140],"execute":[122],"bit-serial":[123],"operations.":[124],"analysis":[126],"reveals":[127],"that":[128,201],"greatly":[130],"benefits":[131],"NNs:":[133],"UPMEM":[135],"provides":[136],"23\u00d7":[137],"high-end":[142],"graphics":[143],"processing":[144,175,189,191],"unit":[145,176,192],"(GPU)":[146],"when":[147],"GPU":[149],"requires":[150],"oversubscription":[152],"general":[155],"matrix\u2013vector":[156],"multiplication":[157],"kernel,":[158],"Mensa":[160],"improves":[161],"efficiency":[163],"throughput":[165],"3.0\u00d7":[167],"3.1\u00d7":[169],"over":[170],"baseline":[172],"Edge":[173],"tensor":[174],"24":[178],"Google":[179],"models,":[182],"SIMDRAM":[185],"outperforms":[186],"central":[188],"unit/graphics":[190],"16.7\u00d7/1.4\u00d7":[194],"binary":[197],"We":[199],"conclude":[200],"ideal":[203],"models":[208],"depends":[209],"on":[210],"model's":[212],"distinct":[213],"attributes,":[214],"due":[215],"inherent":[218],"architectural":[219],"design":[220],"choices.":[221]},"counts_by_year":[{"year":2026,"cited_by_count":13},{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2022-08-31T00:00:00"}