{"id":"https://openalex.org/W4323343906","doi":"https://doi.org/10.14778/3579075.3579077","title":"PetPS: Supporting Huge Embedding Models with Persistent Memory","display_name":"PetPS: Supporting Huge Embedding Models with Persistent Memory","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4323343906","doi":"https://doi.org/10.14778/3579075.3579077"},"language":"en","primary_location":{"id":"doi:10.14778/3579075.3579077","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3579075.3579077","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"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":"Proceedings of the VLDB Endowment","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/A5101139637","display_name":"Minhui Xie","orcid":"https://orcid.org/0000-0001-6684-8336"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Minhui Xie","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052095173","display_name":"Youyou Lu","orcid":"https://orcid.org/0000-0002-6214-5390"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youyou Lu","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100434842","display_name":"Qing Wang","orcid":"https://orcid.org/0000-0002-5526-7154"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qing Wang","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041437245","display_name":"Yangyang Feng","orcid":"https://orcid.org/0000-0003-2317-8330"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yangyang Feng","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059021626","display_name":"Jiaqiang Liu","orcid":"https://orcid.org/0000-0001-5818-7822"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaqiang Liu","raw_affiliation_strings":["Kuaishou"],"affiliations":[{"raw_affiliation_string":"Kuaishou","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046958624","display_name":"Kai Ren","orcid":"https://orcid.org/0000-0002-9312-9705"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kai Ren","raw_affiliation_strings":["Kuaishou"],"affiliations":[{"raw_affiliation_string":"Kuaishou","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101740783","display_name":"Jiwu Shu","orcid":"https://orcid.org/0000-0002-7362-2789"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiwu Shu","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5101139637"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":1.9139,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.88254842,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":"16","issue":"5","first_page":"1013","last_page":"1022"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9973000288009644,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9973000288009644,"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/T11478","display_name":"Caching and Content Delivery","score":0.9969000220298767,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9962000250816345,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7334945201873779},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.6744951605796814},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6693025827407837},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.5984203815460205},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5148448348045349},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.4731132686138153},{"id":"https://openalex.org/keywords/hash-function","display_name":"Hash function","score":0.45597872138023376},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.40692436695098877},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.33339619636535645},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.19782891869544983},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.1534232795238495},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.09753388166427612},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.09040138125419617},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.08829224109649658}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7334945201873779},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.6744951605796814},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6693025827407837},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.5984203815460205},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5148448348045349},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4731132686138153},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.45597872138023376},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.40692436695098877},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.33339619636535645},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.19782891869544983},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.1534232795238495},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.09753388166427612},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.09040138125419617},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.08829224109649658}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14778/3579075.3579077","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3579075.3579077","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"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":"Proceedings of the VLDB Endowment","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4699999988079071,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1985229168","https://openalex.org/W2060393849","https://openalex.org/W2339765813","https://openalex.org/W2433709474","https://openalex.org/W2475334473","https://openalex.org/W2532546572","https://openalex.org/W2604662567","https://openalex.org/W2605350416","https://openalex.org/W2723293840","https://openalex.org/W2734941459","https://openalex.org/W2793768763","https://openalex.org/W2962745591","https://openalex.org/W2964182926","https://openalex.org/W2975438130","https://openalex.org/W2984020950","https://openalex.org/W3010663313","https://openalex.org/W3010969086","https://openalex.org/W3012550338","https://openalex.org/W3016842236","https://openalex.org/W3088708161","https://openalex.org/W3093519337","https://openalex.org/W3100941245","https://openalex.org/W3103616267","https://openalex.org/W3104030692","https://openalex.org/W3130104841","https://openalex.org/W3158307504","https://openalex.org/W4211178602","https://openalex.org/W4214658871","https://openalex.org/W4226328099","https://openalex.org/W4281737296","https://openalex.org/W4282576620","https://openalex.org/W4287186706"],"related_works":["https://openalex.org/W3120961607","https://openalex.org/W3148568549","https://openalex.org/W361036515","https://openalex.org/W2161286015","https://openalex.org/W1648516568","https://openalex.org/W52283896","https://openalex.org/W4211178602","https://openalex.org/W2269474412","https://openalex.org/W4386903460","https://openalex.org/W2800626838"],"abstract_inverted_index":{"Embedding":[0],"models":[1],"are":[2,11],"effective":[3],"for":[4,18,89,115],"learning":[5],"high-dimensional":[6],"sparse":[7],"data.":[8],"Traditionally,":[9],"they":[10],"deployed":[12,186],"in":[13,187],"DRAM":[14],"parameter":[15,87,99,131],"servers":[16],"(PS)":[17],"online":[19,90],"inference":[20],"access.":[21,122],"However,":[22],"the":[23,64,95,126,178],"ever-increasing":[24],"model":[25,117],"capacity":[26,54],"makes":[27],"this":[28],"practice":[29],"suffer":[30],"from":[31],"both":[32],"high":[33,73,105],"storage":[34],"costs":[35],"and":[36,76,144,194],"long":[37],"recovery":[38],"time.":[39],"Rapidly":[40],"developing":[41],"Persistent":[42],"Memory":[43],"(PM)":[44],"offers":[45],"new":[46],"opportunities":[47],"to":[48,51,119,133,135,162],"PSs":[49,163],"owing":[50],"its":[52,61],"large":[53],"at":[55],"low":[56],"costs,":[57],"as":[58,60],"well":[59],"persistence,":[62],"while":[63],"application":[65],"of":[66],"PM":[67,98,106,111,121,143,167],"also":[68],"faces":[69],"two":[70],"challenges":[71],"including":[72],"read":[74],"latency":[75,175],"heavy":[77],"CPU":[78,127,137,147],"burden.":[79],"To":[80,102,124],"provide":[81],"a":[82,110],"low-cost":[83],"but":[84],"still":[85],"high-performance":[86],"service":[88],"inferences,":[91],"we":[92],"introduce":[93],"PetPS,":[94],"first":[96],"production-deployed":[97],"server.":[100],"(1)":[101],"escape":[103],"with":[104,177],"latency,":[107],"PetPS":[108,129,153,183],"introduces":[109],"hash":[112,168],"index":[113],"tailored":[114],"embedding":[116],"workloads,":[118],"minimize":[120],"(2)":[123],"alleviate":[125],"burden,":[128],"offloads":[130],"gathering":[132],"NICs,":[134],"avoid":[136],"stalls":[138],"when":[139],"accessing":[140],"parameters":[141],"on":[142],"thus":[145],"improve":[146],"efficiency.":[148],"Our":[149],"evaluation":[150],"shows":[151],"that":[152,164],"can":[154],"boost":[155],"throughput":[156],"by":[157,198],"1.3":[158],"--":[159,173],"1.7X":[160],"compared":[161],"use":[165],"state-of-the-art":[166],"indexes,":[169],"or":[170],"get":[171],"2.9":[172],"5.5X":[174],"reduction":[176],"same":[179],"throughput.":[180],"Since":[181],"2020,":[182],"has":[184],"been":[185],"Kuaishou,":[188],"one":[189],"world-leading":[190],"short":[191],"video":[192],"company,":[193],"successfully":[195],"reduced":[196],"TCO":[197],"30%":[199],"without":[200],"performance":[201],"degradation.":[202]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-10-10T00:00:00"}