{"id":"https://openalex.org/W4408894144","doi":"https://doi.org/10.1145/3676641.3716011","title":"Past-Future Scheduler for LLM Serving under SLA Guarantees","display_name":"Past-Future Scheduler for LLM Serving under SLA Guarantees","publication_year":2025,"publication_date":"2025-03-27","ids":{"openalex":"https://openalex.org/W4408894144","doi":"https://doi.org/10.1145/3676641.3716011"},"language":"en","primary_location":{"id":"doi:10.1145/3676641.3716011","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3676641.3716011","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2507.10150","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5048414655","display_name":"Ruihao Gong","orcid":"https://orcid.org/0000-0002-6024-7086"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ruihao Gong","raw_affiliation_strings":["Beihang University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073769778","display_name":"Shihao Bai","orcid":"https://orcid.org/0000-0001-6885-8423"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shihao Bai","raw_affiliation_strings":["SenseTime, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"SenseTime, Shanghai, China","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059763565","display_name":"Siyu Wu","orcid":"https://orcid.org/0009-0006-4295-9983"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Siyu Wu","raw_affiliation_strings":["Beihang University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111231988","display_name":"Yiyang Fan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yunqian Fan","raw_affiliation_strings":["SenseTime, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"SenseTime, Shanghai, China","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108503103","display_name":"Zaijun Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zaijun Wang","raw_affiliation_strings":["SenseTime, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"SenseTime, Shanghai, China","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114246744","display_name":"Xiuhong Li","orcid":"https://orcid.org/0000-0002-4896-121X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiuhong Li","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018705589","display_name":"Hailong Yang","orcid":"https://orcid.org/0000-0003-1101-7927"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hailong Yang","raw_affiliation_strings":["Beihang University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024067284","display_name":"Xianglong Liu","orcid":"https://orcid.org/0000-0002-7618-3275"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xianglong Liu","raw_affiliation_strings":["Beihang University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5048414655"],"corresponding_institution_ids":["https://openalex.org/I82880672"],"apc_list":null,"apc_paid":null,"fwci":9.081,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.97464524,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"798","last_page":"813"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10933","display_name":"Real-Time Systems Scheduling","score":0.993399977684021,"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"}},"topics":[{"id":"https://openalex.org/T10933","display_name":"Real-Time Systems Scheduling","score":0.993399977684021,"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"}},{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9800999760627747,"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":0.9789000153541565,"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/computer-science","display_name":"Computer science","score":0.7440711259841919},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.428046852350235},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.375984787940979}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7440711259841919},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.428046852350235},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.375984787940979}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3676641.3716011","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3676641.3716011","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2507.10150","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.10150","pdf_url":"https://arxiv.org/pdf/2507.10150","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2507.10150","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.10150","pdf_url":"https://arxiv.org/pdf/2507.10150","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":7,"referenced_works":["https://openalex.org/W2606555609","https://openalex.org/W2954698171","https://openalex.org/W4285294723","https://openalex.org/W4387321091","https://openalex.org/W4389524393","https://openalex.org/W4400377444","https://openalex.org/W6602430550"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"The":[0,26],"exploration":[1],"and":[2,137,163],"application":[3,99],"of":[4,28,36,40,68,133,153,176],"Large":[5],"Language":[6],"Models":[7],"(LLMs)":[8],"is":[9,217],"thriving.":[10],"To":[11,106],"reduce":[12],"deployment":[13],"costs,":[14],"continuous":[15,29],"batching":[16,30],"has":[17],"become":[18],"an":[19,33],"essential":[20],"feature":[21],"in":[22,47,63,224],"current":[23],"service":[24],"frameworks.":[25],"effectiveness":[27,175],"relies":[31],"on":[32],"accurate":[34],"estimate":[35],"the":[37,45,119,125,130,158,174,177,190,222],"memory":[38,69,121,139],"requirements":[39],"requests.":[41],"However,":[42],"due":[43],"to":[44,54,83,148,172,194,206,220],"diversity":[46],"request":[48,76,134,161],"output":[49,135],"lengths,":[50],"existing":[51,195],"frameworks":[52],"tend":[53],"adopt":[55],"aggressive":[56,196],"or":[57,66,78,197],"conservative":[58,198],"schedulers,":[59,199],"which":[60],"often":[61],"result":[62],"significant":[64],"overestimation":[65],"underestimation":[67],"consumption.":[70],"Consequently,":[71],"they":[72],"suffer":[73],"from":[74],"harmful":[75,164],"evictions":[77],"prolonged":[79],"queuing":[80,162],"times,":[81],"failing":[82],"achieve":[84],"satisfactory":[85],"throughput":[86],"under":[87,213],"strict":[88],"Service":[89],"Level":[90],"Agreement":[91],"(SLA)":[92],"guarantees":[93],"(a.k.a.":[94],"goodput),":[95],"across":[96],"various":[97],"LLM":[98,184],"scenarios":[100],"with":[101,150],"differing":[102],"input-output":[103,154],"length":[104,155],"distributions.":[105],"address":[107],"this":[108],"issue,":[109],"we":[110,180],"propose":[111],"a":[112,182],"novel":[113],"Past-Future":[114,191],"scheduler":[115],"that":[116,188],"precisely":[117],"estimates":[118],"peak":[120],"resources":[122],"required":[123],"by":[124],"running":[126],"batch":[127],"via":[128],"considering":[129],"historical":[131],"distribution":[132],"lengths":[136],"calculating":[138],"occupancy":[140],"at":[141],"each":[142],"future":[143],"time":[144],"point.":[145],"It":[146],"adapts":[147],"applications":[149],"all":[151],"types":[152],"distributions,":[156],"balancing":[157],"trade-off":[159],"between":[160],"evictions,":[165],"thereby":[166],"consistently":[167],"achieving":[168,204],"better":[169],"goodput.":[170],"Furthermore,":[171],"validate":[173],"proposed":[178],"scheduler,":[179],"developed":[181],"high-performance":[183],"serving":[185],"framework,":[186],"LightLLM,":[187],"implements":[189],"scheduler.":[192],"Compared":[193],"LightLLM":[200,216],"demonstrates":[201],"superior":[202],"goodput,":[203],"up":[205],"2-3\u00d7":[207],"higher":[208],"goodput":[209],"than":[210],"other":[211],"schedulers":[212],"heavy":[214],"loads.":[215],"open":[218],"source":[219],"boost":[221],"research":[223],"such":[225],"direction":[226],"(https://github.com/ModelTC/lightllm).":[227]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}