{"id":"https://openalex.org/W7204186536","doi":"https://doi.org/10.48550/arxiv.2608.21719","title":"PowerSlider: Exploiting Phase Asymmetry for LLM Serving under Demand Response","display_name":"PowerSlider: Exploiting Phase Asymmetry for LLM Serving under Demand Response","publication_year":2026,"publication_date":"2026-08-22","ids":{"openalex":"https://openalex.org/W7204186536","doi":"https://doi.org/10.48550/arxiv.2608.21719"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.21719","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21719","pdf_url":"https://arxiv.org/pdf/2608.21719","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.21719","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148096421","display_name":"Yueying Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405261073","display_name":"Cornell Tech","ror":"https://ror.org/04qscbg47","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295","https://openalex.org/I4405261073"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Yueying","raw_affiliation_strings":["Cornell Tech"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell Tech","institution_ids":["https://openalex.org/I4405261073"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062587013","display_name":"Jiayang Chen","orcid":"https://orcid.org/0009-0003-1305-9159","observed_orcids":["https://orcid.org/0009-0003-1305-9159"]},"institutions":[{"id":"https://openalex.org/I4405261073","display_name":"Cornell Tech","ror":"https://ror.org/04qscbg47","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295","https://openalex.org/I4405261073"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen, Jiayang","raw_affiliation_strings":["Cornell Tech"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell Tech","institution_ids":["https://openalex.org/I4405261073"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055208000","display_name":"Yanlun Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405261073","display_name":"Cornell Tech","ror":"https://ror.org/04qscbg47","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295","https://openalex.org/I4405261073"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen, Yuanfan","raw_affiliation_strings":["Cornell Tech"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell Tech","institution_ids":["https://openalex.org/I4405261073"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153473825","display_name":"Leo Han","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405261073","display_name":"Cornell Tech","ror":"https://ror.org/04qscbg47","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295","https://openalex.org/I4405261073"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Han, Leo","raw_affiliation_strings":["Cornell Tech"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell Tech","institution_ids":["https://openalex.org/I4405261073"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037793114","display_name":"Haoran Qiu","orcid":"https://orcid.org/0000-0002-8071-1130","observed_orcids":["https://orcid.org/0000-0002-8071-1130"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qiu, Haoran","raw_affiliation_strings":["Cornell University Microsoft Azure Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell University Microsoft Azure Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045708568","display_name":"Esha Choukse","orcid":"https://orcid.org/0000-0003-0371-5522","observed_orcids":["https://orcid.org/0000-0003-0371-5522"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Choukse, Esha","raw_affiliation_strings":["Cornell University Microsoft Azure Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell University Microsoft Azure Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036529548","display_name":"Rodrigo Fonseca","orcid":"https://orcid.org/0000-0001-9662-2661","observed_orcids":["https://orcid.org/0000-0001-9662-2661"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fonseca, Rodrigo","raw_affiliation_strings":["Cornell University Microsoft Azure Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell University Microsoft Azure Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035704218","display_name":"Udit Gupta","orcid":"https://orcid.org/0000-0002-9118-0961","observed_orcids":["https://orcid.org/0000-0002-9118-0961"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gupta, Udit","raw_affiliation_strings":["Cornell University Microsoft Azure Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cornell University Microsoft Azure Research","institution_ids":["https://openalex.org/I1290206253"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14098","display_name":"Energy Efficiency in Computing","score":0.5259000062942505,"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/T14098","display_name":"Energy Efficiency in Computing","score":0.5259000062942505,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.14470000565052032,"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"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.1289999932050705,"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/large-language-model-inference","display_name":"large language model inference","score":0.9639999866485596},{"id":"https://openalex.org/keywords/demand-response","display_name":"demand response","score":0.9769999980926514},{"id":"https://openalex.org/keywords/dynamic-voltage-and-frequency-scaling","display_name":"dynamic voltage and frequency scaling","score":0.7850000262260437},{"id":"https://openalex.org/keywords/phase-asymmetry","display_name":"phase asymmetry","score":0.6940000057220459},{"id":"https://openalex.org/keywords/service-level-objectives","display_name":"service level objectives","score":0.4959999918937683},{"id":"https://openalex.org/keywords/sglang","display_name":"SGLang","score":0.5979999899864197}],"concepts":[{"id":"https://openalex.org/C94022561","wikidata":"https://www.wikidata.org/wiki/Q1172393","display_name":"Goodput","level":4,"score":0.9639000296592712},{"id":"https://openalex.org/C2779438525","wikidata":"https://www.wikidata.org/wiki/Q5255048","display_name":"Demand response","level":3,"score":0.5674999952316284},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5343000292778015},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.48649999499320984},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4797999858856201},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4731999933719635},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.4341000020503998},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.4068000018596649},{"id":"https://openalex.org/C65155139","wikidata":"https://www.wikidata.org/wiki/Q5380912","display_name":"Envelope (radar)","level":3,"score":0.4052000045776367},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.39910000562667847},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.391400009393692},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.37279999256134033},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.32899999618530273},{"id":"https://openalex.org/C44280652","wikidata":"https://www.wikidata.org/wiki/Q104837","display_name":"Phase (matter)","level":2,"score":0.3278999924659729},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.31529998779296875},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.30309998989105225},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.3005000054836273},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2921999990940094},{"id":"https://openalex.org/C2781007418","wikidata":"https://www.wikidata.org/wiki/Q1456934","display_name":"Capacity planning","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C162932704","wikidata":"https://www.wikidata.org/wiki/Q1058791","display_name":"Synchronizing","level":3,"score":0.28040000796318054},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.27250000834465027},{"id":"https://openalex.org/C196921405","wikidata":"https://www.wikidata.org/wiki/Q786431","display_name":"Online algorithm","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C206658404","wikidata":"https://www.wikidata.org/wiki/Q12725","display_name":"Electricity","level":2,"score":0.2700999975204468},{"id":"https://openalex.org/C2983254600","wikidata":"https://www.wikidata.org/wiki/Q1096907","display_name":"Power grid","level":3,"score":0.2653999924659729},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.2531999945640564}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.21719","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21719","pdf_url":"https://arxiv.org/pdf/2608.21719","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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.2608.21719","doi":"https://doi.org/10.48550/arxiv.2608.21719","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.21719","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.21719","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21719","pdf_url":"https://arxiv.org/pdf/2608.21719","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":[{"score":0.571399986743927,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"},{"score":0.5378999710083008,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"study_designs":[],"awards":[{"id":"https://openalex.org/G1453181110","display_name":null,"funder_award_id":"#2138286","funder_id":"https://openalex.org/F4320337377","funder_display_name":"Office of Advanced Cyberinfrastructure"},{"id":"https://openalex.org/G1912064254","display_name":"Track 4:  Advanced CI Coordination Ecosystem: Monitoring and Measurement Services","funder_award_id":"2137603","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3562386289","display_name":null,"funder_award_id":"#2138296","funder_id":"https://openalex.org/F4320337377","funder_display_name":"Office of Advanced Cyberinfrastructure"},{"id":"https://openalex.org/G4006316252","display_name":"Track 1: ACCESS Resource Allocations Marketplace and Platform Services (RAMPS)","funder_award_id":"2138259","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4010157413","display_name":null,"funder_award_id":"2138259","funder_id":"https://openalex.org/F4320337377","funder_display_name":"Office of Advanced Cyberinfrastructure"},{"id":"https://openalex.org/G4625599001","display_name":"Collaborative Research: National Science Foundation Expeditions in Computing: Carbon Connect -- An Ecosystem for Sustainable Computing","funder_award_id":"2326608","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4676224565","display_name":"ACO: An Open CI Ecosystem to Advance Scientific Discovery (OpenCI)","funder_award_id":"2138296","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G509285653","display_name":"Track 3: COre National Ecosystem for CyberinfrasTructure (CONECT)","funder_award_id":"2138307","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5367127336","display_name":null,"funder_award_id":"#2138307","funder_id":"https://openalex.org/F4320337377","funder_display_name":"Office of Advanced Cyberinfrastructure"},{"id":"https://openalex.org/G5464919783","display_name":null,"funder_award_id":"2137603","funder_id":"https://openalex.org/F4320337377","funder_display_name":"Office of Advanced Cyberinfrastructure"},{"id":"https://openalex.org/G7564846446","display_name":"Collaborative Research: PPoSS: LARGE: Panorama: Integrated Rack-Scale Acceleration for Computational Pangenomics","funder_award_id":"2118709","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8300691027","display_name":"Track 2: Customized Multi-tier Assistance, Training, and Computational Help (MATCH) for End User ACCESS to CI","funder_award_id":"2138286","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"},{"id":"https://openalex.org/F4320337377","display_name":"Office of Advanced Cyberinfrastructure","ror":"https://ror.org/04nh1dc89"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204186536.pdf","grobid_xml":"https://content.openalex.org/works/W7204186536.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W4391045969","https://openalex.org/W4399657778","https://openalex.org/W4401211704","https://openalex.org/W4407310199","https://openalex.org/W4409248734","https://openalex.org/W4414281281","https://openalex.org/W4414760376","https://openalex.org/W4415152247","https://openalex.org/W4415203459"],"related_works":["https://openalex.org/W7172408880","https://openalex.org/W7162036542","https://openalex.org/W7167092193","https://openalex.org/W7125943596","https://openalex.org/W7212287174","https://openalex.org/W7163868207","https://openalex.org/W4414760376","https://openalex.org/W7203826518","https://openalex.org/W7163638510","https://openalex.org/W7077504210"],"abstract_inverted_index":{"AI":[0],"inference":[1],"clusters":[2],"are":[3],"increasingly":[4],"constrained":[5],"by":[6,140],"instantaneous":[7],"power,":[8],"not":[9,53],"just":[10],"energy:":[11],"grid":[12],"operators":[13],"condition":[14],"new":[15,104],"capacity":[16,81],"on":[17,152],"demand":[18],"response,":[19],"imposing":[20],"time-varying":[21],"power":[22,46],"caps.":[23],"Existing":[24],"LLM":[25,50],"serving":[26],"systems":[27],"optimize":[28],"a":[29,54,86,103,126,141,166,197],"static":[30,153],"energy":[31],"objective":[32],"or":[33],"shed":[34],"fixed":[35],"priority":[36],"tiers":[37],"under":[38],"load;":[39],"either":[40],"way,":[41],"goodput":[42,164,195],"collapses":[43],"when":[44,148],"the":[45,96,173,207],"envelope":[47],"moves.":[48],"An":[49],"pipeline":[51],"is":[52],"uniform":[55],"load:":[56],"compute-bound":[57],"prefill":[58],"loses":[59],"throughput":[60],"almost":[61],"linearly":[62],"with":[63,102,157],"GPU":[64],"frequency,":[65],"memory-bound":[66],"answer":[67],"decode":[68],"sustains":[69,161],"it":[70],"down":[71],"to":[72,82,91,189],"$0.57\\times$":[73],"nominal,":[74],"and":[75,122,125,191],"reasoning's":[76],"thinking":[77],"phase":[78],"couples":[79],"KV-cache":[80],"scheduling":[83],"--":[84],"so":[85,101],"cap":[87,137,168],"should":[88],"be":[89],"steered":[90],"where":[92],"each":[93],"watt":[94],"costs":[95],"least":[97],"performance.":[98],"PowerSlider":[99],"does":[100],"Flex":[105],"SLO":[106],"contract":[107],"that":[108,144],"turns":[109],"bounded":[110],"user":[111],"slack":[112],"into":[113],"an":[114],"optimization":[115],"constraint,":[116],"prefill--think--answer":[117],"disaggregation":[118],"exposing":[119],"per-stage":[120],"frequency":[121],"KV":[123],"control,":[124],"Karush--Kuhn--Tucker":[127],"(KKT)":[128],"online":[129,163],"solver":[130],"re-solving":[131],"within":[132,182],"7.7":[133],"ms":[134],"of":[135,175,184],"every":[136,209],"change,":[138],"backed":[139],"consolidated":[142],"fail-safe":[143],"power-gates":[145],"drained":[146],"instances":[147],"DVFS":[149],"bottoms":[150],"out":[151],"power.":[154],"On":[155],"SGLang":[156],"production":[158],"traces,":[159],"\\sys{}":[160],"78.3\\%":[162],"at":[165,203,206],"30\\%":[167],"reduction":[169],"versus":[170],"47.6\\%":[171],"for":[172],"best":[174],"five":[176],"baselines":[177],"($1.64\\times$),":[178],"holds":[179],"latency-critical":[180],"tails":[181],"$1.3\\times$":[183],"nominal":[185],"(baselines:":[186],"$2.3$--$6\\times$,":[187],"up":[188],"$12\\times$),":[190],"delivers":[192],"92\\%":[193],"mean":[194],"through":[196],"replayed":[198],"CAISO":[199],"grid-emergency":[200],"day":[201],"bottoming":[202],"$0.41\\times$":[204],"(54\\%":[205],"trough;":[208],"baseline":[210],"below":[211],"7\\%).":[212]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-26T00:00:00"}