{"id":"https://openalex.org/W4390188278","doi":"https://doi.org/10.1109/hpec58863.2023.10363545","title":"Adaptive Sparse Deep Neural Network Inference on Resource-Constrained Cost-Efficient GPUs","display_name":"Adaptive Sparse Deep Neural Network Inference on Resource-Constrained Cost-Efficient GPUs","publication_year":2023,"publication_date":"2023-09-25","ids":{"openalex":"https://openalex.org/W4390188278","doi":"https://doi.org/10.1109/hpec58863.2023.10363545"},"language":"en","primary_location":{"id":"doi:10.1109/hpec58863.2023.10363545","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec58863.2023.10363545","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE High Performance Extreme Computing Conference (HPEC)","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":"https://openalex.org/A5058494044","display_name":"Ming Dun","orcid":"https://orcid.org/0000-0002-0664-9543"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ming Dun","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106667425","display_name":"Xu Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Zhang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","School of Computer and Control Engineering, UCAS, Beijing, China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"School of Computer and Control Engineering, UCAS, Beijing, China","institution_ids":[]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012125082","display_name":"Huawei Cao","orcid":"https://orcid.org/0000-0003-1176-2521"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huawei Cao","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","Nanjing Institute of InforSuperbahn, Nanjing, China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"Nanjing Institute of InforSuperbahn, Nanjing, China","institution_ids":[]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029752627","display_name":"Yuan Zhang","orcid":"https://orcid.org/0000-0002-7909-9845"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuan Zhang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","School of Computer and Control Engineering, UCAS, Beijing, China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"School of Computer and Control Engineering, UCAS, Beijing, China","institution_ids":[]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032468655","display_name":"Junying Huang","orcid":"https://orcid.org/0000-0001-5845-6965"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junying Huang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023098180","display_name":"Xiaochun Ye","orcid":"https://orcid.org/0000-0003-4598-1685"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaochun Ye","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]},{"raw_affiliation_string":"State Key Lab of Processors, Institute of Computing Technology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210090176"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5058494044"],"corresponding_institution_ids":["https://openalex.org/I4210090176"],"apc_list":null,"apc_paid":null,"fwci":0.7405,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.74040901,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"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":0.9994999766349792,"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":0.9994999766349792,"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":0.9983000159263611,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9965000152587891,"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":0.8624387979507446},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6436344981193542},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.604925274848938},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5630530118942261},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.48130762577056885},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.47503191232681274},{"id":"https://openalex.org/keywords/partition","display_name":"Partition (number theory)","score":0.4310920238494873},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.42413759231567383},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4062982499599457},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3203890323638916},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2788756489753723}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8624387979507446},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6436344981193542},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.604925274848938},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5630530118942261},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.48130762577056885},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.47503191232681274},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.4310920238494873},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42413759231567383},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4062982499599457},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3203890323638916},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2788756489753723},{"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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpec58863.2023.10363545","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec58863.2023.10363545","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1645176854","display_name":null,"funder_award_id":"4232036","funder_id":"https://openalex.org/F4320322919","funder_display_name":"Natural Science Foundation of Beijing Municipality"},{"id":"https://openalex.org/G3112459251","display_name":null,"funder_award_id":"2022YFB4501404","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320322919","display_name":"Natural Science Foundation of Beijing Municipality","ror":null},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W2007339694","https://openalex.org/W2063558363","https://openalex.org/W2735974062","https://openalex.org/W2785452945","https://openalex.org/W2901549770","https://openalex.org/W2914631005","https://openalex.org/W2946417913","https://openalex.org/W2963087201","https://openalex.org/W2963327228","https://openalex.org/W2963854351","https://openalex.org/W2964537638","https://openalex.org/W2973134322","https://openalex.org/W3044913359","https://openalex.org/W3100809954","https://openalex.org/W3100839241","https://openalex.org/W3101584909","https://openalex.org/W3104528661","https://openalex.org/W3112103703","https://openalex.org/W3113443077","https://openalex.org/W3115410382","https://openalex.org/W3125028070","https://openalex.org/W3208099998","https://openalex.org/W4200090124","https://openalex.org/W4280633999","https://openalex.org/W4289300273","https://openalex.org/W4292779060","https://openalex.org/W4293024985","https://openalex.org/W4301361180","https://openalex.org/W4308090217","https://openalex.org/W4308090748","https://openalex.org/W6756592394","https://openalex.org/W6778883912","https://openalex.org/W6787953186"],"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/W1973046741","https://openalex.org/W2983282793","https://openalex.org/W3012895752","https://openalex.org/W3023876411","https://openalex.org/W123152114"],"abstract_inverted_index":{"Sparse":[0],"Deep":[1],"Neural":[2],"Networks":[3],"(SpDNNs)":[4],"has":[5],"gained":[6],"great":[7],"popularity":[8],"and":[9,26,67,89,118,180,195,209,218],"been":[10],"widely":[11],"applied":[12],"in":[13,28,86,99],"various":[14],"machine":[15],"learning":[16],"area.":[17],"Compared":[18],"to":[19,38,49,65,75,132,149,170,212,237],"traditional":[20],"dense":[21],"DNNs,":[22],"the":[23,29,45,73,95,116,126,138,143,152,161,168,172,191,196,203],"unpre-dictable":[24],"irregularity":[25],"sparsity":[27],"sparse":[30],"weight":[31],"matrices":[32],"of":[33,44,163,206],"SpDNNs":[34,51,74,107,175],"make":[35],"them":[36],"difficult":[37],"be":[39,63],"efficiently":[40,150],"parallelled.":[41],"Moreover,":[42],"most":[43],"recent":[46],"advanced":[47],"efforts":[48],"optimize":[50],"are":[52,167],"based":[53],"on":[54,108,156,174,216,241],"high-end":[55],"GPUs":[56,80,109,244],"like":[57],"NVIDIA":[58,220],"V100,":[59],"which":[60],"may":[61],"not":[62],"affordable":[64],"individuals":[66],"smaller":[68],"re-search":[69],"groups.":[70],"However,":[71],"migrating":[72],"those":[76],"cost-efficient":[77],"but":[78],"resource-constrained":[79],"confronts":[81],"enormous":[82],"challenges,":[83],"including":[84],"limitations":[85],"both":[87],"memory":[88,117,134,178],"computing":[90,119,154,183],"resources,":[91],"as":[92,94],"well":[93],"tiresome":[96],"hyper-parameter":[97],"tuning":[98,179],"batch":[100],"parallelism.":[101],"In":[102],"this":[103],"paper,":[104],"we":[105,124,141,166],"accelerate":[106],"with":[110,190,245],"more":[111],"restricted":[112],"resources":[113,155],"through":[114,176],"exploiting":[115],"resources.":[120],"On":[121,137],"one":[122],"hand,":[123,140],"design":[125],"adaptive":[127,177],"memory-aware":[128],"data":[129],"partition":[130],"scheme":[131],"reduce":[133],"consumption":[135],"automatically.":[136],"other":[139],"propose":[142],"Tensor":[144],"core/CUDA":[145],"core":[146,184],"fusion":[147],"mechanism":[148],"utilize":[151],"hetergeneous":[153,182],"modern":[157],"GPU":[158],"architecture.":[159],"To":[160],"best":[162],"our":[164,188,200,227],"knowledge,":[165],"first":[169],"improve":[171],"performance":[173],"utilizing":[181],"concurrency.":[185],"We":[186],"compare":[187],"implementation":[189],"state-of-art":[192],"previous":[193],"champions,":[194],"results":[197],"demonstrate":[198],"that":[199],"work":[201,228],"achieves":[202],"highest":[204],"speedup":[205],"1.35":[207],"x":[208],"1.39x":[210],"compared":[211,236],"2022":[213],"champion":[214,239],"S&Z":[215],"single":[217],"multiple":[219],"Tesla":[221],"T4":[222,248],"GPUs,":[223],"respectively.":[224],"What's":[225],"more,":[226],"can":[229],"reach":[230],"similar":[231],"or":[232],"even":[233],"better":[234],"throughput":[235],"2020":[238],"H&P":[240],"6":[242],"V100":[243],"only":[246],"4":[247],"GPUs.":[249]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":2}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}