{"id":"https://openalex.org/W4414260714","doi":"https://doi.org/10.1145/3767725","title":"Unified and Near-optimal Multi-GPU Cache for Embedding-based Deep Learning","display_name":"Unified and Near-optimal Multi-GPU Cache for Embedding-based Deep Learning","publication_year":2025,"publication_date":"2025-09-13","ids":{"openalex":"https://openalex.org/W4414260714","doi":"https://doi.org/10.1145/3767725"},"language":"en","primary_location":{"id":"doi:10.1145/3767725","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3767725","pdf_url":null,"source":{"id":"https://openalex.org/S193109227","display_name":"ACM Transactions on Computer Systems","issn_l":"0734-2071","issn":["0734-2071","1557-7333"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Computer Systems","raw_type":"journal-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/A5087812146","display_name":"Xiaoniu Song","orcid":"https://orcid.org/0009-0009-1887-2636"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210093034","display_name":"Specific Technologies (United States)","ror":"https://ror.org/00fwkdt40","country_code":"US","type":"company","lineage":["https://openalex.org/I4210093034"]}],"countries":["CN","US"],"is_corresponding":true,"raw_author_name":"Xiaoniu Song","raw_affiliation_strings":["Engineering Research Center for Domain-specific Operating Systems","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai, China","Engineering Research Center for Domain-specific Operating Systems, Shanghai China"],"raw_orcid":"https://orcid.org/0009-0009-1887-2636","affiliations":[{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems, Shanghai China","institution_ids":["https://openalex.org/I4210093034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101543146","display_name":"Rong Chen","orcid":"https://orcid.org/0000-0002-6115-8130"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210093034","display_name":"Specific Technologies (United States)","ror":"https://ror.org/00fwkdt40","country_code":"US","type":"company","lineage":["https://openalex.org/I4210093034"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Rong Chen","raw_affiliation_strings":["Engineering Research Center for Domain-specific Operating Systems","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China","Engineering Research Center for Domain-specific Operating Systems, Shanghai China"],"raw_orcid":"https://orcid.org/0000-0002-6115-8130","affiliations":[{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems, Shanghai China","institution_ids":["https://openalex.org/I4210093034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100775399","display_name":"Haitao Song","orcid":"https://orcid.org/0000-0003-2113-2224"},"institutions":[{"id":"https://openalex.org/I4210093034","display_name":"Specific Technologies (United States)","ror":"https://ror.org/00fwkdt40","country_code":"US","type":"company","lineage":["https://openalex.org/I4210093034"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haitao Song","raw_affiliation_strings":["Engineering Research Center for Domain-specific Operating Systems","Engineering Research Center for Domain-specific Operating Systems, Shanghai China"],"raw_orcid":"https://orcid.org/0000-0003-2113-2224","affiliations":[{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems, Shanghai China","institution_ids":["https://openalex.org/I4210093034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041011260","display_name":"Yiwen Zhang","orcid":"https://orcid.org/0009-0002-0462-1288"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210093034","display_name":"Specific Technologies (United States)","ror":"https://ror.org/00fwkdt40","country_code":"US","type":"company","lineage":["https://openalex.org/I4210093034"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Yiwen Zhang","raw_affiliation_strings":["Engineering Research Center for Domain-specific Operating Systems","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China","Engineering Research Center for Domain-specific Operating Systems, Shanghai China"],"raw_orcid":"https://orcid.org/0009-0002-0462-1288","affiliations":[{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems, Shanghai China","institution_ids":["https://openalex.org/I4210093034"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100406215","display_name":"Haibo Chen","orcid":"https://orcid.org/0000-0002-9720-0361"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210093034","display_name":"Specific Technologies (United States)","ror":"https://ror.org/00fwkdt40","country_code":"US","type":"company","lineage":["https://openalex.org/I4210093034"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Haibo Chen","raw_affiliation_strings":["Engineering Research Center for Domain-specific Operating Systems","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","Engineering Research Center for Domain-specific Operating Systems, Shanghai China","Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China"],"raw_orcid":"https://orcid.org/0000-0002-9720-0361","affiliations":[{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Engineering Research Center for Domain-specific Operating Systems, Shanghai China","institution_ids":["https://openalex.org/I4210093034"]},{"raw_affiliation_string":"Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University, Shanghai China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5087812146"],"corresponding_institution_ids":["https://openalex.org/I183067930","https://openalex.org/I4210093034"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.23091783,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"44","issue":"1","first_page":"1","last_page":"32"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.9987999796867371,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.9987999796867371,"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/T11478","display_name":"Caching and Content Delivery","score":0.9986000061035156,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.9983999729156494,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.7483999729156494},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.6743999719619751},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.6639999747276306},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5575000047683716},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.4916999936103821},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.46709999442100525},{"id":"https://openalex.org/keywords/partition","display_name":"Partition (number theory)","score":0.4629000127315521},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.43470001220703125}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9156000018119812},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.7483999729156494},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.6743999719619751},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6639999747276306},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5575000047683716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5389999747276306},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4916999936103821},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.46709999442100525},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.4629000127315521},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.43470001220703125},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3598000109195709},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3456999957561493},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.31470000743865967},{"id":"https://openalex.org/C48903430","wikidata":"https://www.wikidata.org/wiki/Q491370","display_name":"Graph partition","level":3,"score":0.31130000948905945},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.31119999289512634},{"id":"https://openalex.org/C74197172","wikidata":"https://www.wikidata.org/wiki/Q1195339","display_name":"Directed acyclic graph","level":2,"score":0.3041999936103821},{"id":"https://openalex.org/C38556500","wikidata":"https://www.wikidata.org/wiki/Q13404475","display_name":"Cache algorithms","level":4,"score":0.30410000681877136},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3037000000476837},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3025999963283539},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.27459999918937683},{"id":"https://openalex.org/C12590798","wikidata":"https://www.wikidata.org/wiki/Q3933199","display_name":"Replication (statistics)","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.26429998874664307},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3767725","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3767725","pdf_url":null,"source":{"id":"https://openalex.org/S193109227","display_name":"ACM Transactions on Computer Systems","issn_l":"0734-2071","issn":["0734-2071","1557-7333"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Computer Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G818234559","display_name":null,"funder_award_id":"62272291 and 62132014","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":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W131619556","https://openalex.org/W1986603225","https://openalex.org/W2012643955","https://openalex.org/W2021221734","https://openalex.org/W2033137936","https://openalex.org/W2053359564","https://openalex.org/W2152723376","https://openalex.org/W2157134596","https://openalex.org/W2243416539","https://openalex.org/W2964182926","https://openalex.org/W3008359675","https://openalex.org/W3013235169","https://openalex.org/W3096566397","https://openalex.org/W3152501898","https://openalex.org/W3159953606","https://openalex.org/W3197720002","https://openalex.org/W3198239267","https://openalex.org/W3200211247","https://openalex.org/W3201621211","https://openalex.org/W4214658871","https://openalex.org/W4220807331","https://openalex.org/W4220884018","https://openalex.org/W4226328099","https://openalex.org/W4255033750","https://openalex.org/W4282576620","https://openalex.org/W4282963011","https://openalex.org/W4289646353","https://openalex.org/W4290928031","https://openalex.org/W4296591836","https://openalex.org/W4296591840","https://openalex.org/W4312933868","https://openalex.org/W4318541545","https://openalex.org/W4387302734","https://openalex.org/W4387321090","https://openalex.org/W4396571432","https://openalex.org/W4414034826"],"related_works":[],"abstract_inverted_index":{"This":[0],"article":[1],"presents":[2],"UGache":[3,16,37,63,89,122,148],",":[4],"a":[5,39,60,66],"unified":[6],"multi-GPU":[7],"cache":[8,68],"system":[9],"designed":[10],"for":[11,81],"embedding-based":[12,152],"deep":[13,115,153],"learning":[14,116],"(EmbDL).":[15],"is":[17],"primarily":[18],"motivated":[19],"by":[20,129],"the":[21,78,145],"unique":[22],"characteristics":[23],"of":[24,105,132,147,158],"EmbDL":[25,106],"applications,":[26,107],"namely":[27,108],"read-only":[28],"and":[29,35,56,73,90,98,114,126,134,139],"skewed":[30],"embedding":[31],"accesses":[32],"with":[33,155],"affinity":[34],"predictability.":[36],"introduces":[38],"novel":[40],"factored":[41],"extraction":[42,79],"mechanism":[43],"that":[44,70,121],"avoids":[45],"bandwidth":[46],"congestion":[47],"to":[48,76,137],"fully":[49],"exploit":[50],"high-speed":[51],"cross-GPU":[52],"interconnects":[53],"(e.g.,":[54],"NVLink":[55],"NVSwitch).":[57],"Based":[58],"on":[59,161],"hotness":[61],"metric,":[62],"also":[64],"provides":[65],"near-optimal":[67],"policy":[69],"balances":[71],"local":[72],"remote":[74],"access":[75],"minimize":[77],"time":[80],"diverse":[82],"GPU":[83],"interconnect":[84],"topologies.":[85],"We":[86],"have":[87],"implemented":[88],"integrated":[91],"it":[92],"into":[93],"two":[94,102],"representative":[95],"frameworks,":[96],"TensorFlow":[97],"PyTorch.":[99],"Evaluation":[100],"using":[101],"typical":[103],"types":[104],"graph":[109],"neural":[110],"network":[111],"(GNN)":[112],"training":[113],"recommendation":[117],"(DLR)":[118],"inference,":[119],"shows":[120],"outperforms":[123],"state-of-the-art":[124],"replication":[125],"partition":[127],"designs":[128],"an":[130,156,162],"average":[131],"1.93\u00d7":[133],"1.63\u00d7":[135],"(up":[136],"5.25\u00d7":[138],"3.45\u00d7),":[140],"respectively.":[141],"Furthermore,":[142],"we":[143],"demonstrate":[144],"applicability":[146],"\u2019s":[149],"principle":[150],"beyond":[151],"learning,":[154],"example":[157],"text-to-image":[159],"generation":[160],"inference":[163],"cluster.":[164]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}
