{"id":"https://openalex.org/W4394892775","doi":"https://doi.org/10.1145/3617232.3624849","title":"Proteus: A High-Throughput Inference-Serving System with Accuracy Scaling","display_name":"Proteus: A High-Throughput Inference-Serving System with Accuracy Scaling","publication_year":2024,"publication_date":"2024-04-17","ids":{"openalex":"https://openalex.org/W4394892775","doi":"https://doi.org/10.1145/3617232.3624849"},"language":"en","primary_location":{"id":"doi:10.1145/3617232.3624849","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3617232.3624849","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3617232.3624849","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3617232.3624849","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103133518","display_name":"Sohaib Ahmad","orcid":"https://orcid.org/0009-0008-3892-9460"},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sohaib Ahmad","raw_affiliation_strings":["University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America"],"raw_orcid":"https://orcid.org/0009-0008-3892-9460","affiliations":[{"raw_affiliation_string":"University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America","institution_ids":["https://openalex.org/I24603500"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085489377","display_name":"Hui Guan","orcid":"https://orcid.org/0000-0001-9128-2231"},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hui Guan","raw_affiliation_strings":["University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America"],"raw_orcid":"https://orcid.org/0000-0001-9128-2231","affiliations":[{"raw_affiliation_string":"University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America","institution_ids":["https://openalex.org/I24603500"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007111103","display_name":"Brian D. Friedman","orcid":"https://orcid.org/0009-0003-3534-8787"},"institutions":[{"id":"https://openalex.org/I72090969","display_name":"Nokia (United States)","ror":"https://ror.org/038km2573","country_code":"US","type":"company","lineage":["https://openalex.org/I2738502077","https://openalex.org/I72090969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Brian D. Friedman","raw_affiliation_strings":["Nokia Bell Labs, Murray Hill, New Jersey, USA"],"raw_orcid":"https://orcid.org/0009-0003-3534-8787","affiliations":[{"raw_affiliation_string":"Nokia Bell Labs, Murray Hill, New Jersey, USA","institution_ids":["https://openalex.org/I72090969"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102959987","display_name":"T. Williams","orcid":"https://orcid.org/0009-0008-9675-0928"},"institutions":[{"id":"https://openalex.org/I72090969","display_name":"Nokia (United States)","ror":"https://ror.org/038km2573","country_code":"US","type":"company","lineage":["https://openalex.org/I2738502077","https://openalex.org/I72090969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Thomas Williams","raw_affiliation_strings":["Nokia Bell Labs, Murray Hill, New Jersey, USA"],"raw_orcid":"https://orcid.org/0009-0008-9675-0928","affiliations":[{"raw_affiliation_string":"Nokia Bell Labs, Murray Hill, New Jersey, USA","institution_ids":["https://openalex.org/I72090969"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037538509","display_name":"Ramesh K. Sitaraman","orcid":"https://orcid.org/0000-0003-0558-6875"},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ramesh K. Sitaraman","raw_affiliation_strings":["University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America"],"raw_orcid":"https://orcid.org/0000-0003-0558-6875","affiliations":[{"raw_affiliation_string":"University of Massachusetts, Amherst, Amherst, Massachusetts, United States of America","institution_ids":["https://openalex.org/I24603500"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069403056","display_name":"Thomas Woo","orcid":"https://orcid.org/0009-0003-2028-5533"},"institutions":[{"id":"https://openalex.org/I72090969","display_name":"Nokia (United States)","ror":"https://ror.org/038km2573","country_code":"US","type":"company","lineage":["https://openalex.org/I2738502077","https://openalex.org/I72090969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Thomas Woo","raw_affiliation_strings":["Nokia Bell Labs, Murray Hill, New Jersey, USA"],"raw_orcid":"https://orcid.org/0009-0003-2028-5533","affiliations":[{"raw_affiliation_string":"Nokia Bell Labs, Murray Hill, New Jersey, USA","institution_ids":["https://openalex.org/I72090969"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5103133518"],"corresponding_institution_ids":["https://openalex.org/I24603500"],"apc_list":null,"apc_paid":null,"fwci":5.4664,"has_fulltext":true,"cited_by_count":23,"citation_normalized_percentile":{"value":0.96795576,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"318","last_page":"334"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9986000061035156,"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":0.9986000061035156,"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":0.993399977684021,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9864000082015991,"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":0.8374741077423096},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7493962645530701},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.7098933458328247},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.6791508197784424},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5268204212188721},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.3823263645172119},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3699924051761627},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3588167726993561},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.17501422762870789},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.16996535658836365},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.07283121347427368}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8374741077423096},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7493962645530701},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.7098933458328247},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6791508197784424},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5268204212188721},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3823263645172119},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3699924051761627},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3588167726993561},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.17501422762870789},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.16996535658836365},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.07283121347427368},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","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},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3617232.3624849","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3617232.3624849","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3617232.3624849","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3617232.3624849","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3617232.3624849","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3617232.3624849","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1957757468","display_name":null,"funder_award_id":"2312396","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G2184662362","display_name":null,"funder_award_id":"2220211","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4072858779","display_name":null,"funder_award_id":"CNS-1763617","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6976256797","display_name":"CSR: Medium: Collaborative Research: Foundations of Cache Network Operations for Content Delivery","funder_award_id":"1763617","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7089186217","display_name":null,"funder_award_id":"2106299","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7264624474","display_name":null,"funder_award_id":"2224054","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G884735099","display_name":null,"funder_award_id":"CNS-2106299","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4394892775.pdf","grobid_xml":"https://content.openalex.org/works/W4394892775.grobid-xml"},"referenced_works_count":21,"referenced_works":["https://openalex.org/W2034466462","https://openalex.org/W2194775991","https://openalex.org/W2604319603","https://openalex.org/W2897268228","https://openalex.org/W2944614352","https://openalex.org/W2956612537","https://openalex.org/W2981852735","https://openalex.org/W2982157693","https://openalex.org/W3008618223","https://openalex.org/W3016719260","https://openalex.org/W3026530461","https://openalex.org/W3097411828","https://openalex.org/W3155243801","https://openalex.org/W3158444059","https://openalex.org/W3214037210","https://openalex.org/W4214512541","https://openalex.org/W4214690606","https://openalex.org/W4282577879","https://openalex.org/W4288089799","https://openalex.org/W4318623169","https://openalex.org/W6950310954"],"related_works":["https://openalex.org/W141820298","https://openalex.org/W2049584446","https://openalex.org/W2079781215","https://openalex.org/W4378770497","https://openalex.org/W2136583354","https://openalex.org/W2111238207","https://openalex.org/W4308245303","https://openalex.org/W2014033564","https://openalex.org/W3128807919","https://openalex.org/W3176411177"],"abstract_inverted_index":{"Existing":[0],"machine":[1],"learning":[2],"inference-serving":[3,75,156],"systems":[4],"largely":[5],"rely":[6],"on":[7,117,149,164],"hardware":[8,25,42,60],"scaling":[9,26,79,102],"by":[10,103,176,183],"adding":[11],"more":[12,16],"devices":[13],"or":[14,35],"using":[15],"powerful":[17],"accelerators":[18],"to":[19,39,62,95,109,114,122,126,137,178,187],"handle":[20,63,138],"increasing":[21],"query":[22,65,124,141],"demands.":[23,66],"However,":[24],"might":[27],"not":[28],"be":[29],"feasible":[30],"for":[31],"fixed-size":[32],"edge":[33],"clusters":[34],"private":[36],"clouds":[37],"due":[38],"their":[40],"limited":[41],"resources.":[43],"A":[44],"viable":[45],"alternate":[46],"solution":[47],"is":[48],"accuracy":[49,54,78,101,174],"scaling,":[50],"which":[51],"adapts":[52],"the":[53,70,90,97,150],"of":[55,59,72,100],"ML":[56],"models":[57],"instead":[58],"resources":[61],"varying":[64],"This":[67],"work":[68,93],"studies":[69],"design":[71],"a":[73,132],"high-throughput":[74],"system":[76,157],"with":[77,185],"that":[80,171],"can":[81],"meet":[82],"throughput":[83,192],"requirements":[84],"while":[85,190],"maximizing":[86],"accuracy.":[87],"To":[88],"achieve":[89],"goal,":[91],"this":[92],"proposes":[94,131],"identify":[96],"right":[98],"amount":[99],"jointly":[104],"optimizing":[105],"three":[106],"sub-problems:":[107],"how":[108,113,121],"select":[110],"model":[111],"variants,":[112],"place":[115],"them":[116],"heterogeneous":[118],"devices,":[119],"and":[120,144,160,166,180],"assign":[123],"workloads":[125],"each":[127],"device.":[128],"It":[129],"also":[130],"new":[133],"adaptive":[134],"batching":[135],"algorithm":[136],"variations":[139],"in":[140],"arrival":[142],"times":[143],"minimize":[145],"SLO":[146],"violations.":[147],"Based":[148],"proposed":[151],"techniques,":[152],"we":[153],"build":[154],"an":[155],"called":[158],"Proteus":[159,172],"empirically":[161],"evaluate":[162],"it":[163],"real-world":[165],"synthetic":[167],"traces.":[168],"We":[169],"show":[170],"reduces":[173],"drop":[175],"up":[177],"3\u00d7":[179],"latency":[181],"timeouts":[182],"2--10\u00d7":[184],"respect":[186],"baseline":[188],"schemes,":[189],"meeting":[191],"requirements.":[193]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":17},{"year":2024,"cited_by_count":3}],"updated_date":"2026-04-25T08:17:42.794288","created_date":"2025-10-10T00:00:00"}