{"id":"https://openalex.org/W4221150565","doi":"https://doi.org/10.1109/hpca53966.2022.00019","title":"Hercules: Heterogeneity-Aware Inference Serving for At-Scale Personalized Recommendation","display_name":"Hercules: Heterogeneity-Aware Inference Serving for At-Scale Personalized Recommendation","publication_year":2022,"publication_date":"2022-04-01","ids":{"openalex":"https://openalex.org/W4221150565","doi":"https://doi.org/10.1109/hpca53966.2022.00019"},"language":"en","primary_location":{"id":"doi:10.1109/hpca53966.2022.00019","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca53966.2022.00019","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2203.07424","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Liu Ke","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Liu Ke","raw_affiliation_strings":["Meta"],"affiliations":[{"raw_affiliation_string":"Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035704218","display_name":"Udit Gupta","orcid":"https://orcid.org/0000-0002-9118-0961"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Udit Gupta","raw_affiliation_strings":["Meta"],"affiliations":[{"raw_affiliation_string":"Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018352758","display_name":"Mark Hempstead","orcid":"https://orcid.org/0000-0001-9696-4741"},"institutions":[{"id":"https://openalex.org/I121934306","display_name":"Tufts University","ror":"https://ror.org/05wvpxv85","country_code":"US","type":"education","lineage":["https://openalex.org/I121934306"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mark Hempstead","raw_affiliation_strings":["Tufts University"],"affiliations":[{"raw_affiliation_string":"Tufts University","institution_ids":["https://openalex.org/I121934306"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028220093","display_name":"Carole-Jean Wu","orcid":"https://orcid.org/0000-0002-9032-7239"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Carole-Jean Wu","raw_affiliation_strings":["Meta"],"affiliations":[{"raw_affiliation_string":"Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072539515","display_name":"Hsien-Hsin S. Lee","orcid":"https://orcid.org/0000-0002-8926-8243"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hsien-Hsin S. Lee","raw_affiliation_strings":["Meta"],"affiliations":[{"raw_affiliation_string":"Meta","institution_ids":[]}]},{"author_position":"last","author":{"id":null,"display_name":"Xuan Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I204465549","display_name":"Washington University in St. Louis","ror":"https://ror.org/01yc7t268","country_code":"US","type":"education","lineage":["https://openalex.org/I204465549"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xuan Zhang","raw_affiliation_strings":["Washington University in St. Louis"],"affiliations":[{"raw_affiliation_string":"Washington University in St. Louis","institution_ids":["https://openalex.org/I204465549"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.08561533,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"141","last_page":"154"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13497","display_name":"Hermeneutics and Narrative Identity","score":0.9879000186920166,"subfield":{"id":"https://openalex.org/subfields/1211","display_name":"Philosophy"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T13497","display_name":"Hermeneutics and Narrative Identity","score":0.9879000186920166,"subfield":{"id":"https://openalex.org/subfields/1211","display_name":"Philosophy"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13695","display_name":"Aging, Elder Care, and Social Issues","score":0.9749000072479248,"subfield":{"id":"https://openalex.org/subfields/3600","display_name":"General Health Professions"},"field":{"id":"https://openalex.org/fields/36","display_name":"Health Professions"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T13099","display_name":"Health, Medicine and Society","score":0.95660001039505,"subfield":{"id":"https://openalex.org/subfields/3600","display_name":"General Health Professions"},"field":{"id":"https://openalex.org/fields/36","display_name":"Health Professions"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8312299251556396},{"id":"https://openalex.org/keywords/provisioning","display_name":"Provisioning","score":0.6958049535751343},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.6120008230209351},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5489084720611572},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5453930497169495},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5216032266616821},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.47660842537879944},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.30835336446762085},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.21895188093185425},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.13000380992889404}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8312299251556396},{"id":"https://openalex.org/C172191483","wikidata":"https://www.wikidata.org/wiki/Q1071806","display_name":"Provisioning","level":2,"score":0.6958049535751343},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.6120008230209351},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5489084720611572},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5453930497169495},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5216032266616821},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.47660842537879944},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.30835336446762085},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21895188093185425},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.13000380992889404},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/hpca53966.2022.00019","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca53966.2022.00019","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2203.07424","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2203.07424","pdf_url":"https://arxiv.org/pdf/2203.07424","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2203.07424","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2203.07424","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2203.07424","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2203.07424","pdf_url":"https://arxiv.org/pdf/2203.07424","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4221150565.pdf","grobid_xml":"https://content.openalex.org/works/W4221150565.grobid-xml"},"referenced_works_count":35,"referenced_works":["https://openalex.org/W1977661221","https://openalex.org/W2023214828","https://openalex.org/W2071563913","https://openalex.org/W2102709380","https://openalex.org/W2150139096","https://openalex.org/W2156077332","https://openalex.org/W2160121678","https://openalex.org/W2475334473","https://openalex.org/W2512971201","https://openalex.org/W2604514113","https://openalex.org/W2606722458","https://openalex.org/W2723293840","https://openalex.org/W2794670651","https://openalex.org/W2929502194","https://openalex.org/W2962745591","https://openalex.org/W2969388332","https://openalex.org/W2973172293","https://openalex.org/W2979719709","https://openalex.org/W3012143782","https://openalex.org/W3016842236","https://openalex.org/W3043023836","https://openalex.org/W3043433718","https://openalex.org/W3093701126","https://openalex.org/W3109610142","https://openalex.org/W3157531038","https://openalex.org/W3158702178","https://openalex.org/W3191222816","https://openalex.org/W3197658084","https://openalex.org/W6747917562","https://openalex.org/W6751349269","https://openalex.org/W6753584990","https://openalex.org/W6756363772","https://openalex.org/W6763737044","https://openalex.org/W6786224425","https://openalex.org/W6798149751"],"related_works":["https://openalex.org/W334583","https://openalex.org/W7696661","https://openalex.org/W14152614","https://openalex.org/W17926013","https://openalex.org/W6686550","https://openalex.org/W13559190","https://openalex.org/W8939044","https://openalex.org/W2334562","https://openalex.org/W5211984","https://openalex.org/W18453799"],"abstract_inverted_index":{"Personalized":[0],"recommendation":[1,30,65,124],"is":[2,45],"an":[3,60],"important":[4,46],"class":[5],"of":[6,14,22,28],"deep-learning":[7],"applications":[8],"that":[9,68],"powers":[10],"a":[11,19,42,79,100,164],"large":[12,94],"collection":[13],"internet":[15],"services":[16],"and":[17,39,47,73,86,137,156],"consumes":[18],"considerable":[20],"amount":[21],"datacenter":[23,44],"resources.":[24],"As":[25],"the":[26,93,117,158],"scale":[27],"production-grade":[29],"systems":[31],"continues":[32],"to":[33,106,133,141],"grow,":[34],"optimizing":[35],"their":[36],"serving":[37,67],"performance":[38],"efficiency":[40],"in":[41,139,149],"heterogeneous":[43,75,119],"can":[48],"translate":[49],"into":[50],"infrastructure":[51],"capacity":[52,154],"saving.":[53],"In":[54],"this":[55],"paper,":[56],"we":[57],"propose":[58],"Hercules,":[59],"optimized":[61],"framework":[62],"for":[63,122],"personalized":[64],"inference":[66],"targets":[69],"diverse":[70],"industry-representative":[71],"models":[72],"cloud-scale":[74],"systems.":[76],"Hercules":[77,150],"performs":[78,129],"two-stage":[80],"optimization":[81],"procedure":[82],"—":[83],"offline":[84],"profiling":[85],"online":[87],"serving.":[88],"The":[89,126,145],"first":[90],"stage":[91,128],"searches":[92],"under-explored":[95],"task":[96],"scheduling":[97],"space":[98],"with":[99],"gradient-based":[101],"search":[102],"algorithm":[103],"achieving":[104],"up":[105],"9.0×":[107],"latency-bounded":[108],"throughput":[109],"improvement":[110],"on":[111],"individual":[112],"servers;":[113],"it":[114],"also":[115],"identifies":[116],"optimal":[118],"server":[120],"architecture":[121],"each":[123],"workload.":[125],"second":[127],"heterogeneity-aware":[130],"cluster":[131,147,153],"provisioning":[132],"optimize":[134],"resource":[135],"mapping":[136],"allocation":[138],"response":[140],"fluctuating":[142],"diurnal":[143],"loads.":[144],"proposed":[146],"scheduler":[148],"achieves":[151],"47.7%":[152],"saving":[155],"reduces":[157],"provisioned":[159],"power":[160],"by":[161],"23.7%":[162],"over":[163],"state-of-the-art":[165],"greedy":[166],"scheduler.":[167]},"counts_by_year":[],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2022-04-03T00:00:00"}