{"id":"https://openalex.org/W7224096827","doi":"https://doi.org/10.48550/arxiv.2610.37307","title":"EdgePace: Deadline- and Freshness-Aware LLM Scheduling on Unified-Memory Platforms","display_name":"EdgePace: Deadline- and Freshness-Aware LLM Scheduling on Unified-Memory Platforms","publication_year":2026,"publication_date":"2026-10-27","ids":{"openalex":"https://openalex.org/W7224096827","doi":"https://doi.org/10.48550/arxiv.2610.37307"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.37307","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.37307","pdf_url":"https://arxiv.org/pdf/2610.37307","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","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/2610.37307","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5152088799","display_name":"Teng Mei","orcid":"https://orcid.org/0009-0009-8344-3483","observed_orcids":["https://orcid.org/0009-0009-8344-3483"]},"institutions":[{"id":"https://openalex.org/I4210116924","display_name":"Chinese University of Hong Kong, Shenzhen","ror":"https://ror.org/02d5ks197","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633","https://openalex.org/I180726961","https://openalex.org/I4210116924"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mei, Teng","raw_affiliation_strings":["The Chinese University of Hong Kong , Shenzhen Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong , Shenzhen Shenzhen , China","institution_ids":["https://openalex.org/I4210116924"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5159990658","display_name":"Chengxuan Pei","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Pei, Chengxuan","raw_affiliation_strings":["KAUST Thuwal , Saudi Arabia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAUST Thuwal , Saudi Arabia","institution_ids":["https://openalex.org/I71920554"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160040338","display_name":"Weijun Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Weijun","raw_affiliation_strings":["Institute for AI Industry Research (AIR) , Tsinghua University Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute for AI Industry Research (AIR) , Tsinghua University Beijing , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160056274","display_name":"Jie Wu","orcid":null,"observed_orcids":[]},"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":"Wu, Jie","raw_affiliation_strings":["Fudan University Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University Shanghai , China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042255975","display_name":"Marco Canini","orcid":"https://orcid.org/0000-0002-5051-4283","observed_orcids":["https://orcid.org/0000-0002-5051-4283"]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Canini, Marco","raw_affiliation_strings":["KAUST Thuwal , Saudi Arabia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAUST Thuwal , Saudi Arabia","institution_ids":["https://openalex.org/I71920554"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045928410","display_name":"Yunming Xiao","orcid":"https://orcid.org/0000-0002-4913-4881","observed_orcids":["https://orcid.org/0000-0002-4913-4881"]},"institutions":[{"id":"https://openalex.org/I4210116924","display_name":"Chinese University of Hong Kong, Shenzhen","ror":"https://ror.org/02d5ks197","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633","https://openalex.org/I180726961","https://openalex.org/I4210116924"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiao, Yunming","raw_affiliation_strings":["The Chinese University of Hong Kong , Shenzhen Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong , Shenzhen Shenzhen , China","institution_ids":["https://openalex.org/I4210116924"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"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/T10933","display_name":"Real-Time Systems Scheduling","score":0.7768999934196472,"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":0.7768999934196472,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.16439999639987946,"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":0.044599998742341995,"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-models","display_name":"large language models","score":0.8349999785423279},{"id":"https://openalex.org/keywords/edge-computing","display_name":"edge computing","score":0.574999988079071},{"id":"https://openalex.org/keywords/unified-memory","display_name":"unified memory","score":0.7699999809265137},{"id":"https://openalex.org/keywords/memory-bandwidth-contention","display_name":"memory bandwidth contention","score":0.7919999957084656},{"id":"https://openalex.org/keywords/deadline-aware-scheduling","display_name":"deadline-aware scheduling","score":0.8460000157356262},{"id":"https://openalex.org/keywords/nvidia-jetson","display_name":"NVIDIA Jetson","score":0.8450000286102295},{"id":"https://openalex.org/keywords/deadline-misses","display_name":"deadline misses","score":0.3959999978542328}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8266000151634216},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6638000011444092},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5972999930381775},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5932999849319458},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.5709999799728394},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4855000078678131},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4343999922275543},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.4009999930858612},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3765000104904175},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3425999879837036},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3140999972820282},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.29750001430511475},{"id":"https://openalex.org/C3018397939","wikidata":"https://www.wikidata.org/wiki/Q3644502","display_name":"Open source","level":3,"score":0.2888999879360199},{"id":"https://openalex.org/C174998907","wikidata":"https://www.wikidata.org/wiki/Q357662","display_name":"Work in process","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.28450000286102295},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.27559998631477356},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26809999346733093},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.37307","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.37307","pdf_url":"https://arxiv.org/pdf/2610.37307","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.37307","doi":"https://doi.org/10.48550/arxiv.2610.37307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.37307","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.37307","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.37307","pdf_url":"https://arxiv.org/pdf/2610.37307","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224096827.pdf","grobid_xml":"https://content.openalex.org/works/W7224096827.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7161131589","https://openalex.org/W7213547305","https://openalex.org/W7167512921","https://openalex.org/W7164818176","https://openalex.org/W4412623478","https://openalex.org/W7171389993","https://openalex.org/W4416439776","https://openalex.org/W4398841363","https://openalex.org/W7166527910","https://openalex.org/W7171655919"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"increasingly":[4],"run":[5],"on":[6,128],"edge":[7],"devices":[8],"alongside":[9],"tasks":[10,21],"that":[11,86],"continuously":[12],"process":[13],"sensor":[14],"data.":[15],"In":[16],"many":[17],"applications,":[18],"these":[19],"foreground":[20,42,69,88,105,150],"prepare":[22],"observations":[23,37,52],"as":[24],"inputs":[25],"to":[26,117,145],"the":[27,56],"LLM,":[28],"whose":[29],"outputs":[30,164],"guide":[31],"subsequent":[32],"processing":[33,70,139],"or":[34],"actions.":[35],"The":[36],"must":[38,47],"be":[39],"processed":[40],"by":[41,153,180],"deadlines,":[43],"and":[44,58,102,108,113,122,132,140],"LLM":[45,65,93,159,163,185],"feedback":[46],"arrive":[48,165],"while":[49,90,155,182],"its":[50],"source":[51],"remain":[53,78],"useful.":[54],"Yet":[55],"CPU":[57],"accelerators":[59],"share":[60],"a":[61,83,129],"unified-memory":[62],"system,":[63],"so":[64],"inference":[66,115],"can":[67],"delay":[68],"through":[71],"memory-bandwidth":[72],"contention":[73,118],"even":[74],"when":[75],"compute":[76],"resources":[77],"available.":[79],"We":[80,125],"present":[81],"EdgePace,":[82],"bandwidth-aware":[84],"runtime":[85],"protects":[87],"deadlines":[89],"sustaining":[91],"timely":[92],"feedback.":[94],"EdgePace":[95,127],"creates":[96],"safe":[97],"suspension":[98],"points":[99],"across":[100],"prefill":[101],"decode,":[103],"monitors":[104],"memory":[106],"demand":[107],"stalls":[109],"at":[110],"low":[111],"cost,":[112],"schedules":[114],"according":[116],"risk,":[119],"deadline":[120,151,177],"slack,":[121],"output":[123],"freshness.":[124],"evaluate":[126],"OnePlus":[130,172],"13":[131],"three":[133],"NVIDIA":[134],"Jetson":[135],"platforms":[136],"using":[137],"multi-camera":[138],"robotic":[141],"target":[142],"tracking.":[143],"Relative":[144],"unmanaged":[146],"inference,":[147],"it":[148],"reduces":[149],"misses":[152,178],"24.1%--96.1%":[154],"retaining":[156],"64.9%--98.5%":[157],"of":[158,184],"throughput.":[160],"All":[161],"measured":[162],"within":[166],"their":[167],"task-defined":[168],"freshness":[169],"windows.":[170],"On":[171],"13,":[173],"for":[174],"example,":[175],"camera":[176],"fall":[179],"91.3%":[181],"90.4%":[183],"throughput":[186],"is":[187],"retained.":[188]},"counts_by_year":[],"updated_date":"2026-11-06T12:05:46.655746","created_date":"2026-10-29T00:00:00"}