{"id":"https://openalex.org/W2617031569","doi":"https://doi.org/10.1145/3061639.3062297","title":"Optimizing Memory Efficiency for Convolution Kernels on Kepler GPUs","display_name":"Optimizing Memory Efficiency for Convolution Kernels on Kepler GPUs","publication_year":2017,"publication_date":"2017-06-13","ids":{"openalex":"https://openalex.org/W2617031569","doi":"https://doi.org/10.1145/3061639.3062297","mag":"2617031569"},"language":"en","primary_location":{"id":"doi:10.1145/3061639.3062297","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3061639.3062297","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Design Automation Conference 2017","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":null,"display_name":"Xiaoming Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Xiaoming Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052870302","display_name":"Jianxu Chen","orcid":"https://orcid.org/0000-0002-8500-1357"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jianxu Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060901632","display_name":"Danny Z. Chen","orcid":"https://orcid.org/0000-0001-6565-2884"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Danny Z. Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100600905","display_name":"Xiaobo Sharon Hu","orcid":"https://orcid.org/0000-0002-6636-9738"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaobo Sharon Hu","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I107639228"],"apc_list":null,"apc_paid":null,"fwci":1.0169,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.85116041,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"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":0.9983000159263611,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9977999925613403,"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/computer-science","display_name":"Computer science","score":0.855675458908081},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.7474279999732971},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.6726526021957397},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6626355051994324},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.5620173215866089},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5520869493484497},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.5205846428871155},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.49683764576911926},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4276929199695587},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.4273967146873474},{"id":"https://openalex.org/keywords/high-memory","display_name":"High memory","score":0.4161882698535919},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3466488718986511},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3372926712036133},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2389974296092987},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.22706338763237},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.1674845814704895},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.1290629804134369},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.07721638679504395}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.855675458908081},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.7474279999732971},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.6726526021957397},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6626355051994324},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.5620173215866089},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5520869493484497},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.5205846428871155},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.49683764576911926},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4276929199695587},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4273967146873474},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.4161882698535919},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3466488718986511},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3372926712036133},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2389974296092987},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.22706338763237},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.1674845814704895},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.1290629804134369},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.07721638679504395},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","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},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3061639.3062297","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3061639.3062297","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Design Automation Conference 2017","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"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/W2004683477","https://openalex.org/W2099021415","https://openalex.org/W2102605133","https://openalex.org/W2147800946","https://openalex.org/W2155893237","https://openalex.org/W2163605009","https://openalex.org/W2166524747","https://openalex.org/W2172654076","https://openalex.org/W2178700889","https://openalex.org/W2255738257","https://openalex.org/W2530879419","https://openalex.org/W2618530766","https://openalex.org/W2950094539","https://openalex.org/W2962835968","https://openalex.org/W2963367891","https://openalex.org/W2963751813","https://openalex.org/W6640442106"],"related_works":["https://openalex.org/W4243474409","https://openalex.org/W2993780376","https://openalex.org/W2607256436","https://openalex.org/W2114927422","https://openalex.org/W2528745529","https://openalex.org/W1966777863","https://openalex.org/W3145767355","https://openalex.org/W4200309432","https://openalex.org/W2739266613","https://openalex.org/W2107186825"],"abstract_inverted_index":{"Convolution":[0],"is":[1,68],"a":[2,50,69,75,113,131,139],"fundamental":[3],"operation":[4],"in":[5,24],"many":[6],"applications,":[7],"such":[8],"as":[9],"computer":[10],"vision,":[11],"natural":[12,51],"language":[13],"processing,":[14,16],"image":[15],"etc.":[17],"Recent":[18],"successes":[19],"of":[20,43,64,87,93],"convolutional":[21],"neural":[22],"networks":[23],"various":[25],"deep":[26],"learning":[27],"applications":[28],"put":[29],"even":[30],"higher":[31],"demand":[32],"on":[33,96,149,151],"fast":[34],"convolution.":[35],"The":[36],"high":[37],"computation":[38,90,127],"throughput":[39],"and":[40,89,109,126,138,160,174],"memory":[41,62,84,123],"bandwidth":[42,63],"graphics":[44],"processing":[45],"units":[46],"(GPUs)":[47],"make":[48],"GPUs":[49,65,88,153],"choice":[52],"for":[53,66,105,112,134,142,170],"accelerating":[54],"convolution":[55,67,102],"operations.":[56],"However,":[57],"maximally":[58],"exploiting":[59],"the":[60,80,83,106,110,135,143,166,171,175],"available":[61],"challenging":[70],"task.":[71],"This":[72],"paper":[73],"introduces":[74],"general":[76,107,144,176],"model":[77],"to":[78],"address":[79],"mismatch":[81],"between":[82],"bank":[85],"width":[86,92],"data":[91,147],"threads.":[94],"Based":[95],"this":[97],"model,":[98],"we":[99,129],"develop":[100],"two":[101],"kernels,":[103],"one":[104,117],"case":[108,115,137,173],"other":[111],"special":[114,136,172],"with":[116],"input":[118],"channel.":[119],"By":[120],"carefully":[121],"optimizing":[122],"access":[124],"patterns":[125],"patterns,":[128],"design":[130],"communication-optimized":[132],"kernel":[133,141],"communication-reduced":[140],"case.":[145],"Experimental":[146],"based":[148],"implementations":[150],"Kepler":[152],"show":[154],"that":[155],"our":[156],"kernels":[157],"achieve":[158],"5.16x":[159],"35.5%":[161],"average":[162],"performance":[163],"improvement":[164],"over":[165],"latest":[167],"cuDNN":[168],"library,":[169],"case,":[177],"respectively.":[178]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":4},{"year":2018,"cited_by_count":2},{"year":2017,"cited_by_count":1}],"updated_date":"2026-04-28T14:05:53.105641","created_date":"2025-10-10T00:00:00"}