{"id":"https://openalex.org/W4308090463","doi":"https://doi.org/10.1109/hpec55821.2022.9926408","title":"Im2win: Memory Efficient Convolution On SIMD Architectures","display_name":"Im2win: Memory Efficient Convolution On SIMD Architectures","publication_year":2022,"publication_date":"2022-09-19","ids":{"openalex":"https://openalex.org/W4308090463","doi":"https://doi.org/10.1109/hpec55821.2022.9926408"},"language":"en","primary_location":{"id":"doi:10.1109/hpec55821.2022.9926408","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec55821.2022.9926408","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2306.14320","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100670324","display_name":"Shuai Lu","orcid":"https://orcid.org/0009-0009-5808-9478"},"institutions":[{"id":"https://openalex.org/I927504317","display_name":"Nanchang Hangkong University","ror":"https://ror.org/0369pvp92","country_code":"CN","type":"education","lineage":["https://openalex.org/I927504317"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuai Lu","raw_affiliation_strings":["Nanchang Hangkong University,Nanchang,China","Nanchang Hangkong University, Nanchang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanchang Hangkong University,Nanchang,China","institution_ids":["https://openalex.org/I927504317"]},{"raw_affiliation_string":"Nanchang Hangkong University, Nanchang, China","institution_ids":["https://openalex.org/I927504317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058286058","display_name":"Jun Chu","orcid":"https://orcid.org/0009-0006-9643-8059"},"institutions":[{"id":"https://openalex.org/I927504317","display_name":"Nanchang Hangkong University","ror":"https://ror.org/0369pvp92","country_code":"CN","type":"education","lineage":["https://openalex.org/I927504317"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Chu","raw_affiliation_strings":["Nanchang Hangkong University,Nanchang,China","Nanchang Hangkong University, Nanchang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanchang Hangkong University,Nanchang,China","institution_ids":["https://openalex.org/I927504317"]},{"raw_affiliation_string":"Nanchang Hangkong University, Nanchang, China","institution_ids":["https://openalex.org/I927504317"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5054943576","display_name":"Tony Liu","orcid":"https://orcid.org/0000-0003-3980-9803"},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]},{"id":"https://openalex.org/I58610484","display_name":"Seattle University","ror":"https://ror.org/02jqc0m91","country_code":"US","type":"education","lineage":["https://openalex.org/I58610484"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xu T. Liu","raw_affiliation_strings":["University of Washington,Seattle,USA","University of Washington, Seattle, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington,Seattle,USA","institution_ids":["https://openalex.org/I201448701","https://openalex.org/I58610484"]},{"raw_affiliation_string":"University of Washington, Seattle, USA","institution_ids":["https://openalex.org/I201448701","https://openalex.org/I58610484"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.7107,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":{"value":0.70911477,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9955999851226807,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9690999984741211,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8079192638397217},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7078821659088135},{"id":"https://openalex.org/keywords/transformation","display_name":"Transformation (genetics)","score":0.5876676440238953},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5258577466011047},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.49344292283058167},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.48513445258140564},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.482920378446579},{"id":"https://openalex.org/keywords/simd","display_name":"SIMD","score":0.47393542528152466},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.45247119665145874},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.44290074706077576},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.4306184649467468},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3486899733543396},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.15986546874046326},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.14810502529144287}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8079192638397217},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7078821659088135},{"id":"https://openalex.org/C204241405","wikidata":"https://www.wikidata.org/wiki/Q461499","display_name":"Transformation (genetics)","level":3,"score":0.5876676440238953},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5258577466011047},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.49344292283058167},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.48513445258140564},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.482920378446579},{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.47393542528152466},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.45247119665145874},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.44290074706077576},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.4306184649467468},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3486899733543396},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.15986546874046326},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.14810502529144287},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/hpec55821.2022.9926408","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec55821.2022.9926408","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2306.14320","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.14320","pdf_url":"https://arxiv.org/pdf/2306.14320","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2306.14320","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.14320","pdf_url":"https://arxiv.org/pdf/2306.14320","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2586291459","display_name":null,"funder_award_id":"62162045","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4308090463.pdf"},"referenced_works_count":33,"referenced_works":["https://openalex.org/W1513554765","https://openalex.org/W1530262073","https://openalex.org/W1686810756","https://openalex.org/W1983572666","https://openalex.org/W2002257715","https://openalex.org/W2155893237","https://openalex.org/W2164797883","https://openalex.org/W2194775991","https://openalex.org/W2516525699","https://openalex.org/W2604319603","https://openalex.org/W2670151302","https://openalex.org/W2754249189","https://openalex.org/W2767476316","https://openalex.org/W2804777790","https://openalex.org/W2883780447","https://openalex.org/W2914462824","https://openalex.org/W2950094539","https://openalex.org/W2951894856","https://openalex.org/W2953384591","https://openalex.org/W4205546691","https://openalex.org/W4244254628","https://openalex.org/W4254299145","https://openalex.org/W4293763691","https://openalex.org/W4295312788","https://openalex.org/W6630903078","https://openalex.org/W6631660994","https://openalex.org/W6637373629","https://openalex.org/W6713134421","https://openalex.org/W6740373327","https://openalex.org/W6744651773","https://openalex.org/W6746082227","https://openalex.org/W6752390237","https://openalex.org/W6766978945"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2099629705","https://openalex.org/W3157543420","https://openalex.org/W4221139464","https://openalex.org/W4289827464"],"abstract_inverted_index":{"Convolution":[0],"is":[1,13],"the":[2,16,51,74,129,143,158,185,194,214],"most":[3],"expensive":[4],"operation":[5],"among":[6],"neural":[7,20,77],"network":[8,78],"operations,":[9],"thus":[10,72,155],"its":[11],"performance":[12,18,211],"critical":[14],"to":[15,57,65,93,190,193,204,213],"overall":[17],"of":[19,76,121,128],"networks.":[21],"Commonly":[22],"used":[23],"convolution":[24,31,167,196,216],"approaches,":[25],"including":[26,172],"general":[27],"matrix":[28],"multiplication":[29],"(GEMM)-based":[30],"and":[32,132,151,154,201,207,217],"direct":[33],"convolution,":[34],"rely":[35],"on":[36,81,199],"im2col":[37,52],"for":[38],"data":[39,45,53,68,87,111,152,220],"transformation":[40,46,54,69,88,112],"or":[41,123],"do":[42],"not":[43,66,85,218],"use":[44],"at":[47,58,70],"all,":[48,71],"respectively.":[49,222],"However,":[50],"can":[55],"lead":[56],"least":[59],"2":[60],"x":[61],"memory":[62,95,149,159,186],"footprint":[63],"compared":[64,192,212],"using":[67,86,219],"limiting":[73],"size":[75],"models":[79],"running":[80],"memory-limited":[82],"systems.":[83],"Meanwhile,":[84],"usually":[89],"performs":[90],"poorly":[91],"due":[92],"nonconsecutive":[94],"access":[96,150],"although":[97],"it":[98],"consumes":[99],"less":[100],"memory.":[101],"To":[102],"solve":[103],"those":[104],"problems,":[105],"we":[106,162],"propose":[107,163],"a":[108,119,140,164],"new":[109],"memory-efficient":[110],"algorithm,":[113],"called":[114],"im2win.":[115],"This":[116],"algorithm":[117,168,183],"refactorizes":[118],"row":[120,141],"square":[122],"rectangle":[124],"dot":[125],"product":[126],"windows":[127,138],"input":[130],"image":[131],"flattens":[133],"unique":[134],"elements":[135],"within":[136],"these":[137],"into":[139],"in":[142,210],"output":[144],"tensor,":[145],"which":[146],"enables":[147],"consecutive":[148],"reuse,":[153],"greatly":[156],"reduces":[157,184],"overhead.":[160],"Furthermore,":[161],"high-performance":[165],"im2win-based":[166],"with":[169],"various":[170],"optimizations,":[171],"vectorization,":[173],"loop":[174],"reordering,":[175],"etc.":[176],"Our":[177],"experimental":[178],"results":[179],"show":[180],"that":[181],"our":[182],"overhead":[187],"by":[188],"average":[189,203],"41.6%":[191],"PyTorch's":[195],"implementation":[197],"based":[198],"im2col,":[200],"achieves":[202],"3.6":[205],"\u00d7":[206],"5.3\u00d7":[208],"speedup":[209],"im2col-based":[215],"transformation,":[221]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}