{"id":"https://openalex.org/W4200579775","doi":"https://doi.org/10.1109/iccad51958.2021.9643503","title":"ScaleDNN: Data Movement Aware DNN Training on Multi-GPU","display_name":"ScaleDNN: Data Movement Aware DNN Training on Multi-GPU","publication_year":2021,"publication_date":"2021-11-01","ids":{"openalex":"https://openalex.org/W4200579775","doi":"https://doi.org/10.1109/iccad51958.2021.9643503"},"language":"en","primary_location":{"id":"doi:10.1109/iccad51958.2021.9643503","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643503","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5068210433","display_name":"Weizheng Xu","orcid":"https://orcid.org/0009-0001-4512-2058"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Weizheng Xu","raw_affiliation_strings":["University of Pittsburgh"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pittsburgh","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112474431","display_name":"Ashutosh Pattnaik","orcid":"https://orcid.org/0000-0003-0367-5989"},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ashutosh Pattnaik","raw_affiliation_strings":["Penn State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Penn State University","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007300551","display_name":"Geng Yuan","orcid":"https://orcid.org/0000-0001-9844-992X"},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Geng Yuan","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100651384","display_name":"Yanzhi Wang","orcid":"https://orcid.org/0000-0002-3024-7990"},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Yanzhi Wang","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026996875","display_name":"Youtao Zhang","orcid":"https://orcid.org/0000-0001-8425-8743"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Youtao Zhang","raw_affiliation_strings":["University of Pittsburgh"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pittsburgh","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087859795","display_name":"Xulong Tang","orcid":"https://orcid.org/0000-0002-3385-2053"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xulong Tang","raw_affiliation_strings":["University of Pittsburgh"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pittsburgh","institution_ids":["https://openalex.org/I170201317"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.1909,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.54211518,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"2","issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":1.0,"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":1.0,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9954000115394592,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9925000071525574,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8931446075439453},{"id":"https://openalex.org/keywords/pci-express","display_name":"PCI Express","score":0.8379455208778381},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.620465099811554},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5836696624755859},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5597650408744812},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.55027836561203},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.48555538058280945},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.34185129404067993},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.27009671926498413},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1297118067741394},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1014355719089508}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8931446075439453},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.8379455208778381},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.620465099811554},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5836696624755859},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5597650408744812},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.55027836561203},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.48555538058280945},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.34185129404067993},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.27009671926498413},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1297118067741394},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1014355719089508},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","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/iccad51958.2021.9643503","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643503","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1332542428","display_name":"RTML: Large: Efficient and Adaptive Real-Time Learning for Next Generation Wireless Systems","funder_award_id":"1937500","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":55,"referenced_works":["https://openalex.org/W1563795667","https://openalex.org/W1598866093","https://openalex.org/W1821462560","https://openalex.org/W2007511350","https://openalex.org/W2060393849","https://openalex.org/W2084392717","https://openalex.org/W2119144962","https://openalex.org/W2130306094","https://openalex.org/W2131528105","https://openalex.org/W2339765813","https://openalex.org/W2550821151","https://openalex.org/W2551176409","https://openalex.org/W2604319603","https://openalex.org/W2622263826","https://openalex.org/W2766789999","https://openalex.org/W2767239597","https://openalex.org/W2770398148","https://openalex.org/W2775776326","https://openalex.org/W2777078856","https://openalex.org/W2799197246","https://openalex.org/W2809533219","https://openalex.org/W2896457183","https://openalex.org/W2899244816","https://openalex.org/W2900365206","https://openalex.org/W2926767350","https://openalex.org/W2952899695","https://openalex.org/W2963323070","https://openalex.org/W2963327228","https://openalex.org/W2963935227","https://openalex.org/W2969388332","https://openalex.org/W2969766737","https://openalex.org/W3004495293","https://openalex.org/W3011751313","https://openalex.org/W3037847693","https://openalex.org/W3046502253","https://openalex.org/W3088514010","https://openalex.org/W3092260380","https://openalex.org/W3096403968","https://openalex.org/W3129831491","https://openalex.org/W3170887803","https://openalex.org/W4288093768","https://openalex.org/W4288346545","https://openalex.org/W4295312788","https://openalex.org/W4301239768","https://openalex.org/W6652367586","https://openalex.org/W6665801690","https://openalex.org/W6679349572","https://openalex.org/W6746719585","https://openalex.org/W6747195836","https://openalex.org/W6753584990","https://openalex.org/W6758034089","https://openalex.org/W6761333114","https://openalex.org/W6766978945","https://openalex.org/W6769475105","https://openalex.org/W6784425352"],"related_works":["https://openalex.org/W2317245370","https://openalex.org/W4249323025","https://openalex.org/W198851386","https://openalex.org/W2030310580","https://openalex.org/W947442053","https://openalex.org/W1980160788","https://openalex.org/W2148915962","https://openalex.org/W2283866686","https://openalex.org/W4287182096","https://openalex.org/W27867058"],"abstract_inverted_index":{"Training":[0],"Deep":[1],"Neural":[2],"Networks":[3],"(DNNs)":[4],"models":[5],"is":[6,50,180],"a":[7,70,79],"time-consuming":[8],"process":[9],"that":[10,57,81,149],"requires":[11],"immense":[12],"amount":[13,53],"of":[14,43,54,93,158,187],"data":[15,55,68,109,121,133,139,165],"and":[16,83,86,97,111,126,129],"computation.":[17,144],"To":[18],"this":[19,49,74],"end,":[20],"GPUs":[21,64],"are":[22],"widely":[23],"adopted":[24],"to":[25,116,153,182],"accelerate":[26],"the":[27,31,38,41,51,59,63,66,120,124,132,138,142],"training":[28,33,89],"process.":[29],"However,":[30],"delivered":[32],"performance":[34],"rarely":[35],"scales":[36],"with":[37,65,141,155,184],"increase":[39],"in":[40,69],"number":[42],"GPUs.":[44],"The":[45,171],"major":[46],"reason":[47],"behind":[48],"large":[52],"movement":[56,122,134],"prevents":[58],"system":[60,179],"from":[61],"providing":[62],"required":[67],"timely":[71],"fashion.":[72],"In":[73],"paper,":[75],"we":[76],"propose":[77],"ScaleDNN,":[78],"framework":[80],"systematically":[82],"comprehensively":[84],"investigates":[85],"optimizes":[87],"data-parallel":[88],"on":[90,167,176],"two":[91],"types":[92],"multi-GPU":[94,169,178],"systems":[95],"(PCIe-based":[96],"NVLink-based).":[98],"Specifically,":[99],"ScaleDNN":[100,150],"performs:":[101],"i)":[102],"CPU-centric":[103],"input":[104],"batch":[105],"splitting,":[106],"ii)":[107],"mini-batch":[108],"pre-loading,":[110],"iii)":[112],"model":[113],"parameter":[114],"compression":[115],"effectively":[117],"a)":[118],"reduce":[119],"between":[123],"CPU":[125],"multiple":[127],"GPUs,":[128],"b)":[130],"hide":[131],"overheads":[135],"by":[136],"overlapping":[137],"transfer":[140],"GPU":[143],"Our":[145],"experimental":[146],"results":[147],"show":[148],"achieves":[151],"up":[152,181],"39.38%,":[154],"an":[156,185],"average":[157,186],"17.96%":[159],"execution":[160,173],"time":[161,174],"saving":[162],"over":[163],"modern":[164],"parallelism":[166],"PCIe-based":[168],"system.":[170],"corresponding":[172],"reduction":[175],"NVLink-based":[177],"19.20%":[183],"10.26%.":[188]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1}],"updated_date":"2026-08-01T09:00:35.917206","created_date":"2025-10-10T00:00:00"}