{"id":"https://openalex.org/W4295935130","doi":"https://doi.org/10.1109/tc.2022.3207134","title":"Future Scaling of Memory Hierarchy for Tensor Cores and Eliminating Redundant Shared Memory Traffic Using Inter-Warp Multicasting","display_name":"Future Scaling of Memory Hierarchy for Tensor Cores and Eliminating Redundant Shared Memory Traffic Using Inter-Warp Multicasting","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W4295935130","doi":"https://doi.org/10.1109/tc.2022.3207134"},"language":"en","primary_location":{"id":"doi:10.1109/tc.2022.3207134","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tc.2022.3207134","pdf_url":"https://ieeexplore.ieee.org/ielx7/12/4358213/09893362.pdf","source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://ieeexplore.ieee.org/ielx7/12/4358213/09893362.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5025001772","display_name":"Sunjung Lee","orcid":"https://orcid.org/0000-0002-5177-0916"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Sunjung Lee","raw_affiliation_strings":["Department of Intelligence and Information, Seoul National University (SNU), South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Intelligence and Information, Seoul National University (SNU), South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065665585","display_name":"Seunghwan Hwang","orcid":"https://orcid.org/0000-0002-7146-1926"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seunghwan Hwang","raw_affiliation_strings":["Department of Intelligence and Information, Seoul National University (SNU), South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Intelligence and Information, Seoul National University (SNU), South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109756929","display_name":"Michael Jaemin Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Michael Jaemin Kim","raw_affiliation_strings":["Department of Intelligence and Information, Seoul National University (SNU), South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Intelligence and Information, Seoul National University (SNU), South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032839104","display_name":"Jaewan Choi","orcid":"https://orcid.org/0000-0003-2447-4369"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaewan Choi","raw_affiliation_strings":["Department of Intelligence and Information, Seoul National University (SNU), South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Intelligence and Information, Seoul National University (SNU), South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5078262826","display_name":"Jung Ho Ahn","orcid":"https://orcid.org/0000-0003-1733-1394"},"institutions":[{"id":"https://openalex.org/I4210120602","display_name":"Advanced Institute of Convergence Technology","ror":"https://ror.org/01w62yz22","country_code":"KR","type":"facility","lineage":["https://openalex.org/I139264467","https://openalex.org/I4210120602"]},{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jung Ho Ahn","raw_affiliation_strings":["Department of Intelligence and Information, Seoul National University (SNU), South Korea","Interdisciplinary Program in Artificial Intelligence, the Research Institute for Convergence Science, Institute of Computer Technology, SNU, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Intelligence and Information, Seoul National University (SNU), South Korea","institution_ids":["https://openalex.org/I139264467"]},{"raw_affiliation_string":"Interdisciplinary Program in Artificial Intelligence, the Research Institute for Convergence Science, Institute of Computer Technology, SNU, South Korea","institution_ids":["https://openalex.org/I4210120602"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5025001772"],"corresponding_institution_ids":["https://openalex.org/I139264467"],"apc_list":null,"apc_paid":null,"fwci":0.403,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.59379776,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"12"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994000196456909,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994000196456909,"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"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994000196456909,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9987999796867371,"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/computer-science","display_name":"Computer science","score":0.8444452285766602},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7215681076049805},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.7130453586578369},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.7074329257011414},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.658210039138794},{"id":"https://openalex.org/keywords/tucker-decomposition","display_name":"Tucker decomposition","score":0.5705626606941223},{"id":"https://openalex.org/keywords/shared-memory","display_name":"Shared memory","score":0.4919540584087372},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.48426076769828796},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.4663592576980591},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4195650517940521},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.418364554643631},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.3401781916618347},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.10292592644691467},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.09640613198280334},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.07949885725975037}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8444452285766602},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7215681076049805},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.7130453586578369},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.7074329257011414},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.658210039138794},{"id":"https://openalex.org/C42704193","wikidata":"https://www.wikidata.org/wiki/Q7851097","display_name":"Tucker decomposition","level":4,"score":0.5705626606941223},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.4919540584087372},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.48426076769828796},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.4663592576980591},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4195650517940521},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.418364554643631},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.3401781916618347},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.10292592644691467},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.09640613198280334},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.07949885725975037},{"id":"https://openalex.org/C2986737658","wikidata":"https://www.wikidata.org/wiki/Q30103009","display_name":"Tensor decomposition","level":3,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tc.2022.3207134","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tc.2022.3207134","pdf_url":"https://ieeexplore.ieee.org/ielx7/12/4358213/09893362.pdf","source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1109/tc.2022.3207134","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tc.2022.3207134","pdf_url":"https://ieeexplore.ieee.org/ielx7/12/4358213/09893362.pdf","source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.6499999761581421,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4295935130.pdf","grobid_xml":"https://content.openalex.org/works/W4295935130.grobid-xml"},"referenced_works_count":50,"referenced_works":["https://openalex.org/W2041811089","https://openalex.org/W2079038734","https://openalex.org/W2097117768","https://openalex.org/W2100913437","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2235724720","https://openalex.org/W2441593292","https://openalex.org/W2566040696","https://openalex.org/W2625052843","https://openalex.org/W2625200202","https://openalex.org/W2626935846","https://openalex.org/W2736230459","https://openalex.org/W2796649226","https://openalex.org/W2798724095","https://openalex.org/W2896457183","https://openalex.org/W2910737925","https://openalex.org/W2945580137","https://openalex.org/W2953264125","https://openalex.org/W2955425717","https://openalex.org/W2973727699","https://openalex.org/W2979340153","https://openalex.org/W2992192853","https://openalex.org/W3004853346","https://openalex.org/W3012406028","https://openalex.org/W3015468748","https://openalex.org/W3035160371","https://openalex.org/W3036692157","https://openalex.org/W3089469538","https://openalex.org/W3102510044","https://openalex.org/W3105696640","https://openalex.org/W3130554079","https://openalex.org/W3142875983","https://openalex.org/W3159607817","https://openalex.org/W3190062760","https://openalex.org/W4287704453","https://openalex.org/W4302296459","https://openalex.org/W4309793872","https://openalex.org/W6637151318","https://openalex.org/W6684191040","https://openalex.org/W6739106156","https://openalex.org/W6750448596","https://openalex.org/W6755207826","https://openalex.org/W6762718338","https://openalex.org/W6767997687","https://openalex.org/W6771656455","https://openalex.org/W6776048684","https://openalex.org/W6779898197","https://openalex.org/W6781533629","https://openalex.org/W6783944145"],"related_works":["https://openalex.org/W2594155508","https://openalex.org/W1997955449","https://openalex.org/W2031026393","https://openalex.org/W2755428707","https://openalex.org/W1990309876","https://openalex.org/W164750744","https://openalex.org/W2950263839","https://openalex.org/W4294538768","https://openalex.org/W4295935130","https://openalex.org/W2150669850"],"abstract_inverted_index":{"The":[0],"CUDA":[1,91,119],"core":[2,43,92,96,109,130,145],"of":[3,9,31,89,106,132,196],"NVIDIA":[4,38,148],"GPUs":[5],"had":[6],"been":[7],"one":[8],"the":[10,41,60,69,74,78,90,94,102,107,118,169,179,187],"most":[11],"efficient":[12],"computation":[13],"units":[14],"for":[15],"parallel":[16],"computing.":[17],"However,":[18,47],"recent":[19,45],"rapid":[20],"developments":[21],"in":[22,44,51,73,82,97,117,193],"deep":[23],"neural":[24],"networks":[25],"demand":[26],"an":[27,157],"even":[28],"higher":[29,112],"level":[30],"computational":[32,52,79],"performance.":[33],"To":[34],"meet":[35],"this":[36,64],"requirement,":[37],"has":[39],"introduced":[40,146],"Tensor":[42,95,108,129,144],"generations.":[46],"their":[48],"impressive":[49],"enhancements":[50],"performance":[53,80,104,188],"have":[54],"newly":[55],"brought":[56],"high":[57],"pressure":[58,182],"on":[59,191],"memory":[61,71,75,113,125,165],"hierarchy.":[62],"In":[63],"paper,":[65],"first":[66],"we":[67,99,122,139,155],"identify":[68],"required":[70],"bandwidth":[72,114,126,181],"hierarchy":[76],"as":[77],"increases":[81],"actual":[83],"GPU":[84,151],"hardware.":[85],"Through":[86],"a":[87,141,150],"comparison":[88],"and":[93,136,185,198],"V100,":[98,133],"find":[100],"that":[101,116,161,175],"tremendous":[103],"increase":[105],"requires":[110],"much":[111],"than":[115],"core.":[120],"Moreover,":[121],"thoroughly":[123],"investigate":[124],"requirement":[127],"over":[128],"generations":[131],"RTX":[134],"TITAN,":[135],"A100.":[137],"Lastly,":[138],"analyze":[140],"hypothetical":[142],"next-generation":[143],"by":[147,183,189],"through":[149,153],"simulation,":[152],"which":[154],"propose":[156],"inter-warp":[158,176],"multicasting":[159,177],"microarchitecture":[160],"reduces":[162,178],"redundant":[163],"shared":[164],"(SMEM)":[166],"traffic":[167],"during":[168],"GEMM":[170],"process.":[171],"Our":[172],"evaluation":[173],"shows":[174],"SMEM":[180],"33%":[184],"improves":[186],"19%":[190],"average":[192],"all":[194],"layers":[195],"ResNet-152":[197],"BERT-Large.":[199]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}