{"id":"https://openalex.org/W3095498018","doi":"https://doi.org/10.1109/cluster49012.2020.00050","title":"Optimizing GPU Memory Transactions for Convolution Operations","display_name":"Optimizing GPU Memory Transactions for Convolution Operations","alternate_titles":[],"publication_year":2020,"publication_date":"2020-09-01","ids":{"openalex":"https://openalex.org/W3095498018","doi":"https://doi.org/10.1109/cluster49012.2020.00050","mag":"3095498018"},"language":"en","primary_location":{"id":"doi:10.1109/cluster49012.2020.00050","doi":"https://doi.org/10.1109/cluster49012.2020.00050","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cluster49012.2020.00050","pdf_url":null,"source":{"id":"https://openalex.org/S4210207155","display_name":"IEEE International Conference on Cluster Computing (CLUSTER)","issn_l":"1552-5244","issn":["1552-5244","2168-9253"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","jufo-1","norway-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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE International Conference on Cluster Computing (CLUSTER)","raw_type":"proceedings-article"},"parent_works":[],"lineage":["https://openalex.org/W3095498018"],"child_works_count":0,"type":"conference-paper","is_traditional":true,"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/A5082302042","display_name":"Gangzhao Lu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Gangzhao Lu","raw_affiliation_strings":["Computer Science and Technology Harbin Institute of Technology, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science and Technology Harbin Institute of Technology, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085221201","display_name":"Weizhe Zhang","orcid":"https://orcid.org/0000-0003-4783-876X","observed_orcids":["https://orcid.org/0000-0003-4783-876X"]},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weizhe Zhang","raw_affiliation_strings":["Computer Science and Technology Harbin Institute of Technology, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science and Technology Harbin Institute of Technology, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100401045","display_name":"Zheng Wang","orcid":"https://orcid.org/0000-0001-6157-0662","observed_orcids":["https://orcid.org/0000-0001-6157-0662","https://orcid.org/0009-0001-7858-6238"]},"institutions":[{"id":"https://openalex.org/I130828816","display_name":"University of Leeds","ror":"https://ror.org/024mrxd33","country_code":"GB","type":"education","lineage":["https://openalex.org/I130828816"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Zheng Wang","raw_affiliation_strings":["School of Computing University of Leeds, United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing University of Leeds, United Kingdom","institution_ids":["https://openalex.org/I130828816"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.8071,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.69153083,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":86,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"399","last_page":"403"},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9980000257492065,"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":0.9980000257492065,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0010999999940395355,"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/T10481","display_name":"Computer Graphics and Visualization Techniques","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1704","display_name":"Computer Graphics and Computer-Aided Design"},"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/convolution-operation","display_name":"convolution operation","score":0.6600000262260437},{"id":"https://openalex.org/keywords/memory-access-optimization","display_name":"memory access optimization","score":0.7390000224113464},{"id":"https://openalex.org/keywords/data-locality","display_name":"data locality","score":0.7039999961853027},{"id":"https://openalex.org/keywords/deep-neural-network-inference","display_name":"deep neural network inference","score":0.6449999809265137},{"id":"https://openalex.org/keywords/nvidia-gpus","display_name":"NVIDIA GPUs","score":0.6290000081062317},{"id":"https://openalex.org/keywords/cudnn","display_name":"cuDNN","score":0.9340000152587891}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8511531352996826},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.726801872253418},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.7027537822723389},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5337859988212585},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.5074438452720642},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.5029377341270447},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.5027806758880615},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40933287143707275},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.38513636589050293},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.31416869163513184},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.15836861729621887},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.136185884475708},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09147337079048157},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.0793566107749939},{"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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/cluster49012.2020.00050","doi":"https://doi.org/10.1109/cluster49012.2020.00050","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cluster49012.2020.00050","pdf_url":null,"source":{"id":"https://openalex.org/S4210207155","display_name":"IEEE International Conference on Cluster Computing (CLUSTER)","issn_l":"1552-5244","issn":["1552-5244","2168-9253"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","jufo-1","norway-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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE International Conference on Cluster Computing (CLUSTER)","raw_type":"proceedings-article"},{"id":"pmh:oai:eprints.whiterose.ac.uk:164433","doi":null,"is_oa":false,"landing_page_url":"https://eprints.whiterose.ac.uk/164433/1/cluster.pdf","pdf_url":"https://eprints.whiterose.ac.uk/164433/1/cluster.pdf","source":{"id":"https://openalex.org/S4306400854","display_name":"White Rose Research Online (University of Leeds, The University of Sheffield, University of York)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2800616092","host_organization_name":"White Rose University Consortium","host_organization_lineage":["https://openalex.org/I2800616092"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Proceedings Paper"}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W1530262073","https://openalex.org/W1667652561","https://openalex.org/W1686810756","https://openalex.org/W1789336918","https://openalex.org/W1922123711","https://openalex.org/W1977901491","https://openalex.org/W2072912074","https://openalex.org/W2097117768","https://openalex.org/W2155893237","https://openalex.org/W2163605009","https://openalex.org/W2172654076","https://openalex.org/W2194775991","https://openalex.org/W2605739168","https://openalex.org/W2670151302","https://openalex.org/W2786374423","https://openalex.org/W2963340555"],"related_works":["https://openalex.org/W2617031569","https://openalex.org/W2806428272","https://openalex.org/W2072912074","https://openalex.org/W4386106891","https://openalex.org/W4409811354","https://openalex.org/W7169733183","https://openalex.org/W2951810231","https://openalex.org/W4401361028","https://openalex.org/W4407692956","https://openalex.org/W4391903527"],"abstract_inverted_index":{"Convolution":[0],"computation":[1],"is":[2,12],"a":[3,55,137],"common":[4],"operation":[5],"in":[6],"deep":[7],"neural":[8],"networks":[9],"(DNNs)":[10],"and":[11,20,89,158,203],"often":[13,36],"responsible":[14],"for":[15,24,49,62,82,111],"performance":[16,50,175,221],"bottlenecks":[17],"during":[18],"training":[19],"inferencing.":[21],"Existing":[22],"approaches":[23],"accelerating":[25],"convolution":[26,63,83,93,119],"operations":[27,81,84,120],"aim":[28],"to":[29,58,75,103,139,144,156,189,201,235],"reduce":[30,76],"computational":[31],"complexity.":[32],"However,":[33],"these":[34],"strategies":[35],"increase":[37],"the":[38,77,87,96,105,109,113,122,131,146,177,192,223,239],"memory":[39,43,60,80,116],"footprint":[40],"with":[41,133],"extra":[42],"accesses,":[44],"thereby":[45],"leaving":[46],"much":[47],"room":[48],"improvement.":[51],"This":[52],"paper":[53],"presents":[54],"novel":[56],"approach":[57,70,155,171,200,216],"optimize":[59],"access":[61],"operations,":[64],"specifically":[65],"targeting":[66],"GPU":[67],"execution.":[68],"Our":[69],"leverages":[71],"two":[72],"optimization":[73],"techniques":[74],"number":[78,114],"of":[79,108,115,130,136,149,195,242],"performed":[85],"on":[86,95,121,162,207],"width":[88,97],"height":[90,123],"dimensions.":[91],"For":[92,118,167,182,212,228],"computations":[94],"dimension,":[98,124],"we":[99,125,186,232],"exploit":[100],"shuffle":[101],"instructions":[102],"exchange":[104],"overlapped":[106,128],"columns":[107],"input":[110,132],"reducing":[112],"transactions.":[117],"multiply":[126],"each":[127],"row":[129,150],"multiple":[134,141],"rows":[135],"filter":[138],"compute":[140],"output":[142],"elements":[143],"improve":[145],"data":[147],"locality":[148],"elements.":[151],"We":[152,197],"apply":[153,198],"our":[154,170,199,215],"2D":[157,160,168,184,202,205,213,230],"multi-channel":[159,183,204,229],"convolutions":[161,206],"an":[163,208],"NVIDIA":[164,209],"2080Ti":[165,210],"GPU.":[166,211],"convolution,":[169,214],"delivers":[172,217],"over":[173,191,218,238],"faster":[174,220],"than":[176,222],"state-of-the-art":[178,224],"image":[179,225],"processing":[180,226],"libraries.":[181,227],"convolutions,":[185,231],"obtain":[187,233],"up":[188,234],"speedups":[190,237],"quickest":[193,240],"algorithm":[194,241],"cuDNN.":[196,243],"2\u00d7":[219],"1.3\u00d7":[236]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W3095498018","counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":1}],"updated_date":"2026-11-18T13:11:19.462570","created_date":"2020-11-09T00:00:00"}
