{"id":"https://openalex.org/W4402897342","doi":"https://doi.org/10.1109/iwqos61813.2024.10682873","title":"Optimizing Inference Quality with SmartNIC for Recommendation System","display_name":"Optimizing Inference Quality with SmartNIC for Recommendation System","publication_year":2024,"publication_date":"2024-06-19","ids":{"openalex":"https://openalex.org/W4402897342","doi":"https://doi.org/10.1109/iwqos61813.2024.10682873"},"language":"en","primary_location":{"id":"doi:10.1109/iwqos61813.2024.10682873","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iwqos61813.2024.10682873","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/ACM 32nd International Symposium on Quality of Service (IWQoS)","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/A5047603167","display_name":"Ruixin Shi","orcid":"https://orcid.org/0000-0003-2231-1275"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ruixin Shi","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101788338","display_name":"Ming Yan","orcid":"https://orcid.org/0000-0002-1176-0238"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ming Yan","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100600532","display_name":"Jie Wu","orcid":"https://orcid.org/0000-0003-0404-0707"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jie Wu","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5047603167"],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":1.5464,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.87339162,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.6826000213623047,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10203","display_name":"Recommender Systems and Techniques","score":0.6826000213623047,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.7433831095695496},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6638562083244324},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.5209818482398987},{"id":"https://openalex.org/keywords/recommender-system","display_name":"Recommender system","score":0.5019698143005371},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3317454755306244},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.2812095284461975}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7433831095695496},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6638562083244324},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.5209818482398987},{"id":"https://openalex.org/C557471498","wikidata":"https://www.wikidata.org/wiki/Q554950","display_name":"Recommender system","level":2,"score":0.5019698143005371},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3317454755306244},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2812095284461975},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iwqos61813.2024.10682873","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iwqos61813.2024.10682873","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/ACM 32nd International Symposium on Quality of Service (IWQoS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2475334473","https://openalex.org/W2531409750","https://openalex.org/W2604662567","https://openalex.org/W2734941459","https://openalex.org/W2896457183","https://openalex.org/W2947737663","https://openalex.org/W3197720002","https://openalex.org/W3200211247","https://openalex.org/W3204524224","https://openalex.org/W4200392450","https://openalex.org/W4220884018","https://openalex.org/W4224903494","https://openalex.org/W4224952058","https://openalex.org/W4226328099","https://openalex.org/W4296415418","https://openalex.org/W4296591817","https://openalex.org/W4296591836","https://openalex.org/W4310744178","https://openalex.org/W4318148006","https://openalex.org/W4381327166","https://openalex.org/W6755207826","https://openalex.org/W6763737044","https://openalex.org/W6774806506","https://openalex.org/W6843273680","https://openalex.org/W6847189266"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W4390273403","https://openalex.org/W4386781444","https://openalex.org/W2150182025","https://openalex.org/W3092950680","https://openalex.org/W3197542405","https://openalex.org/W2056712470","https://openalex.org/W3125580266","https://openalex.org/W4317039510"],"abstract_inverted_index":{"Embedding-based":[0],"recommendation":[1,24],"systems":[2],"are":[3],"now":[4],"widely":[5],"used":[6],"to":[7,55,106,133,148,158],"recommend":[8],"content":[9],"for":[10,42,48],"users,":[11],"and":[12,19,32,45,66,96,113,124,153],"have":[13],"strict":[14],"requirements":[15],"on":[16,39,104,122],"their":[17],"latency":[18,155],"throughput.":[20],"However,":[21],"the":[22,33,63,80,110,119,134,141],"latest":[23],"models":[25],"often":[26,36],"exceed":[27],"GPU":[28,43,68],"HBM":[29],"memory":[30],"capacity,":[31],"system":[34],"is":[35],"deployed":[37],"separately":[38],"computing":[40],"nodes":[41],"calculating":[44],"Parameter":[46],"Servers":[47],"embedding":[49,84,162],"tables\u2019":[50],"storage.":[51],"This":[52],"architecture":[53],"leads":[54],"a":[56,88],"significant":[57],"amount":[58],"of":[59,83,92,143,160],"network":[60,81],"I/O":[61,82],"during":[62],"inference":[64,76,166],"process":[65],"reduces":[67],"utilization.In":[69],"this":[70],"paper,":[71],"we":[72],"propose":[73],"SmartEmb,":[74],"an":[75],"framework":[77],"that":[78,131],"accelerates":[79],"table":[85],"lookups":[86],"through":[87],"specialized":[89],"control":[90,102],"plane":[91],"task":[93],"reordering,":[94],"prefetching":[95],"cache":[97],"management.":[98],"We":[99,117],"offload":[100],"these":[101],"planes":[103],"SmartNIC":[105],"avoid":[107],"contention":[108],"with":[109],"host":[111],"CPU":[112],"gain":[114],"better":[115],"performance.":[116,127],"implemented":[118],"SmartEmb":[120,138],"prototype":[121],"BlueField-2":[123],"evaluated":[125],"its":[126],"Our":[128],"evaluation":[129],"demonstrates":[130],"compared":[132],"Nvidia":[135],"HugeCTR":[136],"HPS,":[137],"can":[139],"improve":[140],"quality":[142],"service":[144],"by":[145,156],"achieving":[146],"up":[147,157],"217%":[149],"improvement":[150],"in":[151,165],"throughput":[152],"reducing":[154],"190%":[159],"overall":[161],"layer":[163],"look-ups":[164],"scenarios.":[167]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-10-10T00:00:00"}