{"id":"https://openalex.org/W7133492435","doi":"https://doi.org/10.1109/hpca68181.2026.11408493","title":"PASCAL: A Phase-Aware Scheduling Algorithm for Serving Reasoning-based Large Language Models","display_name":"PASCAL: A Phase-Aware Scheduling Algorithm for Serving Reasoning-based Large Language Models","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7133492435","doi":"https://doi.org/10.1109/hpca68181.2026.11408493"},"language":null,"primary_location":{"id":"doi:10.1109/hpca68181.2026.11408493","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408493","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","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":null,"display_name":"Eunyeong Cho","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Eunyeong Cho","raw_affiliation_strings":["KAIST,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125995115","display_name":"Jehyeon Bang","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jehyeon Bang","raw_affiliation_strings":["KAIST,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081950817","display_name":"Ranggi Hwang","orcid":"https://orcid.org/0000-0003-2343-586X"},"institutions":[{"id":"https://openalex.org/I48566637","display_name":"Ulsan National Institute of Science and Technology","ror":"https://ror.org/017cjz748","country_code":"KR","type":"education","lineage":["https://openalex.org/I48566637"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Ranggi Hwang","raw_affiliation_strings":["UNIST,South Korea"],"affiliations":[{"raw_affiliation_string":"UNIST,South Korea","institution_ids":["https://openalex.org/I48566637"]}]},{"author_position":"last","author":{"id":null,"display_name":"Minsoo Rhu","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Minsoo Rhu","raw_affiliation_strings":["KAIST,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I157485424"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.44009146,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"16"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.4957999885082245,"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"}},"topics":[{"id":"https://openalex.org/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.4957999885082245,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.07699999958276749,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.07360000163316727,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.42010000348091125},{"id":"https://openalex.org/keywords/fair-share-scheduling","display_name":"Fair-share scheduling","score":0.29170000553131104},{"id":"https://openalex.org/keywords/job-shop-scheduling","display_name":"Job shop scheduling","score":0.26170000433921814},{"id":"https://openalex.org/keywords/rate-monotonic-scheduling","display_name":"Rate-monotonic scheduling","score":0.2556000053882599}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7153000235557556},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.42010000348091125},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3968000113964081},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.32199999690055847},{"id":"https://openalex.org/C31689143","wikidata":"https://www.wikidata.org/wiki/Q733809","display_name":"Fair-share scheduling","level":3,"score":0.29170000553131104},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2827000021934509},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28209999203681946},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.26170000433921814},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2612999975681305},{"id":"https://openalex.org/C127456818","wikidata":"https://www.wikidata.org/wiki/Q238879","display_name":"Rate-monotonic scheduling","level":4,"score":0.2556000053882599},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2554999887943268},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca68181.2026.11408493","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408493","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G8076470811","display_name":null,"funder_award_id":"SRFC-IT2402-03","funder_id":"https://openalex.org/F4320332195","funder_display_name":"Samsung"},{"id":"https://openalex.org/G8534604288","display_name":null,"funder_award_id":"RS-202400438851","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"}],"funders":[{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W4387321091","https://openalex.org/W4398774961","https://openalex.org/W4401211590","https://openalex.org/W4401211704","https://openalex.org/W4402671659","https://openalex.org/W4404386015","https://openalex.org/W4404954309","https://openalex.org/W4406266041","https://openalex.org/W4408894144","https://openalex.org/W4409248734","https://openalex.org/W4409248836","https://openalex.org/W4411119426","https://openalex.org/W4411486537","https://openalex.org/W7133209886"],"related_works":[],"abstract_inverted_index":{"The":[0],"emergence":[1],"of":[2,115],"reasoning-based":[3,119],"LLMs":[4],"leveraging":[5],"Chain-of-Thought":[6],"(CoT)":[7],"inference":[8],"introduces":[9],"new":[10],"serving":[11,27],"challenges,":[12],"as":[13],"their":[14],"extended":[15],"reasoning":[16,33,54],"phases":[17],"delay":[18],"user-visible":[19],"output":[20],"and":[21,34,62,80,90],"inflate":[22],"Time-To-First-Token":[23],"(TTFT).":[24],"Existing":[25],"LLM":[26,120],"frameworks":[28],"fail":[29],"to":[30,38,55,67,87,103],"distinguish":[31],"between":[32],"answering":[35,66,108],"phases,":[36],"leading":[37],"performance":[39],"degradation":[40],"under":[41],"GPU":[42],"memory":[43],"constraints.":[44],"We":[45],"present":[46],"PASCAL,":[47],"a":[48],"phase-aware":[49],"scheduling":[50,117],"algorithm":[51],"that":[52],"prioritizes":[53],"reduce":[56,91],"TTFT":[57,100],"while":[58,106],"using":[59,95],"controlled":[60],"preemption":[61],"token":[63],"pacing":[64],"during":[65],"preserve":[68],"Quality-of-Experience":[69],"(QoE).":[70],"Our":[71],"hierarchical":[72],"scheduler":[73],"combines":[74],"instance-level":[75],"placement":[76],"with":[77],"intrainstance":[78],"execution":[79],"enables":[81],"dynamic":[82],"migration":[83],"at":[84],"phase":[85,109],"boundaries":[86],"balance":[88],"load":[89],"interference.":[92],"Across":[93],"benchmarks":[94],"DeepSeek-R1-Distill-Qwen-32B,":[96],"Pascal":[97],"reduces":[98],"tail":[99],"by":[101],"up":[102],"72":[104],"%":[105],"maintaining":[107],"SLO":[110],"attainment,":[111],"demonstrating":[112],"the":[113],"importance":[114],"phaseaware":[116],"for":[118],"deployment.":[121]},"counts_by_year":[],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2026-02-16T00:00:00"}