{"id":"https://openalex.org/W7140202736","doi":"https://doi.org/10.1109/access.2026.3677249","title":"Context-Aware Autoscaling for Cost-Efficient Large Language Model Inference With Prefix Cache Integration","display_name":"Context-Aware Autoscaling for Cost-Efficient Large Language Model Inference With Prefix Cache Integration","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140202736","doi":"https://doi.org/10.1109/access.2026.3677249"},"language":"en","primary_location":{"id":"doi:10.1109/access.2026.3677249","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3677249","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/access.2026.3677249","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5117533685","display_name":"Seyed Hossein Ahmadpanah","orcid":"https://orcid.org/0009-0007-8460-2074"},"institutions":[{"id":"https://openalex.org/I136830121","display_name":"Islamic Azad University South Tehran Branch","ror":"https://ror.org/02xc21a77","country_code":"IR","type":"education","lineage":["https://openalex.org/I110525433","https://openalex.org/I136830121"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Seyed Hossein Ahmadpanah","raw_affiliation_strings":["Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran"],"raw_orcid":"https://orcid.org/0009-0007-8460-2074","affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran","institution_ids":["https://openalex.org/I136830121"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041757079","display_name":"Amir Sahafi","orcid":"https://orcid.org/0000-0002-6555-670X"},"institutions":[{"id":"https://openalex.org/I136830121","display_name":"Islamic Azad University South Tehran Branch","ror":"https://ror.org/02xc21a77","country_code":"IR","type":"education","lineage":["https://openalex.org/I110525433","https://openalex.org/I136830121"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Amir Sahafi","raw_affiliation_strings":["Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran"],"raw_orcid":"https://orcid.org/0000-0002-6555-670X","affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran","institution_ids":["https://openalex.org/I136830121"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077723176","display_name":"Seyed Hossein Erfani","orcid":"https://orcid.org/0000-0002-7893-4191"},"institutions":[{"id":"https://openalex.org/I136830121","display_name":"Islamic Azad University South Tehran Branch","ror":"https://ror.org/02xc21a77","country_code":"IR","type":"education","lineage":["https://openalex.org/I110525433","https://openalex.org/I136830121"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Seyed Hossein Erfani","raw_affiliation_strings":["Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, ST.C., Islamic Azad University, Tehran, Iran","institution_ids":["https://openalex.org/I136830121"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I136830121"],"apc_list":{"value":1850,"currency":"USD","value_usd":1850},"apc_paid":{"value":1850,"currency":"USD","value_usd":1850},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.43013642,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"14","issue":null,"first_page":"46024","last_page":"46035"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.21619999408721924,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.21619999408721924,"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/T14347","display_name":"Big Data and Digital Economy","score":0.1193000003695488,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.09049999713897705,"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/cache","display_name":"Cache","score":0.6003000140190125},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5131999850273132},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.487199991941452},{"id":"https://openalex.org/keywords/remote-direct-memory-access","display_name":"Remote direct memory access","score":0.46959999203681946},{"id":"https://openalex.org/keywords/prefix","display_name":"Prefix","score":0.4165000021457672},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.40130001306533813},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.3928999900817871},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.3849000036716461},{"id":"https://openalex.org/keywords/locality-of-reference","display_name":"Locality of reference","score":0.38350000977516174},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.36660000681877136}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8939999938011169},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.6003000140190125},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5159000158309937},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5131999850273132},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.487199991941452},{"id":"https://openalex.org/C130795937","wikidata":"https://www.wikidata.org/wiki/Q2561570","display_name":"Remote direct memory access","level":2,"score":0.46959999203681946},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.4255000054836273},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.4165000021457672},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.40130001306533813},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.3928999900817871},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.3849000036716461},{"id":"https://openalex.org/C27602214","wikidata":"https://www.wikidata.org/wiki/Q1868547","display_name":"Locality of reference","level":3,"score":0.38350000977516174},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.36660000681877136},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.35010001063346863},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.34549999237060547},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.34470000863075256},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.33959999680519104},{"id":"https://openalex.org/C121854251","wikidata":"https://www.wikidata.org/wiki/Q62932","display_name":"Elasticity (physics)","level":2,"score":0.3382999897003174},{"id":"https://openalex.org/C38556500","wikidata":"https://www.wikidata.org/wiki/Q13404475","display_name":"Cache algorithms","level":4,"score":0.3278000056743622},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3215000033378601},{"id":"https://openalex.org/C2780609101","wikidata":"https://www.wikidata.org/wiki/Q17156588","display_name":"Resource management (computing)","level":2,"score":0.3059000074863434},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.30379998683929443},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.29760000109672546},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C165435473","wikidata":"https://www.wikidata.org/wiki/Q1509884","display_name":"Padding","level":2,"score":0.2890999913215637},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C2775973920","wikidata":"https://www.wikidata.org/wiki/Q3252726","display_name":"Selection algorithm","level":3,"score":0.2775999903678894},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.26750001311302185},{"id":"https://openalex.org/C53833338","wikidata":"https://www.wikidata.org/wiki/Q1061424","display_name":"Context switch","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C190290938","wikidata":"https://www.wikidata.org/wiki/Q387015","display_name":"Trie","level":3,"score":0.2590000033378601},{"id":"https://openalex.org/C72108876","wikidata":"https://www.wikidata.org/wiki/Q844565","display_name":"Transaction processing","level":3,"score":0.2533000111579895},{"id":"https://openalex.org/C2779370713","wikidata":"https://www.wikidata.org/wiki/Q357554","display_name":"Load management","level":2,"score":0.25049999356269836}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/access.2026.3677249","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3677249","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:769a775f91fc47468edf069fe09b0692","is_oa":true,"landing_page_url":"https://doaj.org/article/769a775f91fc47468edf069fe09b0692","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Access, Vol 14, Pp 46024-46035 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1109/access.2026.3677249","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3677249","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Although":[0],"granular":[1],"resource":[2,105],"management":[3],"has":[4],"been":[5],"made":[6],"possible":[7],"by":[8,80,115,145,212],"the":[9,74,89,180],"architectural":[10],"shift":[11],"to":[12,26,129,151,157,171,215],"Prefill-Decode":[13],"(PD)":[14],"disaggregation":[15],"in":[16,59,93,219],"Large":[17],"Language":[18],"Model":[19],"(LLM)":[20],"serving,":[21],"it":[22,193],"is":[23],"still":[24],"difficult":[25],"maintain":[27],"strict":[28],"Service":[29],"Level":[30],"Objectives":[31],"(SLOs)":[32],"under":[33],"bursty":[34],"traffic.":[35],"Modern":[36],"autoscaling":[37,101],"frameworks,":[38],"like":[39],"TokenScale,":[40],"balance":[41],"resources":[42],"using":[43,196],"fine-grained":[44],"\"velocity\"":[45],"metrics,":[46],"but":[47],"they":[48,63,86],"have":[49],"a":[50,68,99,121,166],"significant":[51],"drawback:cache":[52],"blindness.":[53],"These":[54],"systems":[55],"over-provision":[56],"during":[57,176],"spikes":[58],"high-locality":[60,220],"traffic":[61],"because":[62],"model":[64],"instance":[65],"throughput":[66,77],"as":[67],"static":[69],"hardware":[70],"constant,":[71],"which":[72,140],"ignores":[73],"enormous":[75],"effective":[76],"gains":[78],"offered":[79],"prefix":[81],"caching":[82],"(e.g.,":[83],"RadixAttention).":[84],"Furthermore,":[85],"often":[87],"overlook":[88],"cost-efficiency":[90],"opportunities":[91],"inherent":[92],"heterogeneous":[94],"GPU":[95,209],"clusters.We":[96],"present":[97],"AdaptiveScale,":[98],"context-aware":[100],"framework":[102],"that":[103,124,168,206],"connects":[104],"elasticity":[106],"and":[107,162,192,201],"memory":[108],"locality.":[109],"Three":[110],"new":[111],"mechanisms":[112],"are":[113],"suggested":[114],"AdaptiveScale:":[116],"(1)Effective":[117],"Token":[118],"Velocity":[119],"(Veff),":[120],"dynamic":[122],"metric":[123],"incorporates":[125],"real-time":[126],"Radix-tree":[127],"telemetry":[128],"amplify":[130],"scaling":[131],"logic":[132],"based":[133],"on":[134,190],"context":[135],"reuse;":[136],"(2)Heterogeneity-Aware":[137],"Tiered":[138],"Scheduling,":[139],"optimizes":[141],"convertible":[142],"decoder":[143],"selection":[144],"pinning":[146],"cost-effective":[147],"nodes":[148],"(like":[149,160],"L40S)":[150],"decoding":[152],"while":[153,222],"routing":[154],"prefill":[155],"bursts":[156],"compute-dense":[158],"GPUs":[159],"H100);":[161],"(3)Elastic":[163],"State":[164],"Preservation,":[165],"mechanism":[167],"uses":[169],"RDMA":[170],"zero-copy":[172],"offload":[173],"KV-cache":[174],"states":[175],"role-switching,":[177],"thereby":[178],"removing":[179],"latency":[181],"penalties":[182],"associated":[183],"with":[184],"cache":[185],"eviction.":[186],"AdaptiveScale":[187,207],"was":[188,194],"implemented":[189],"vLLM,":[191],"assessed":[195],"production":[197],"traces":[198],"from":[199],"OpenAI":[200],"Azure.":[202],"Our":[203],"results":[204],"demonstrate":[205],"reduces":[208],"operational":[210],"costs":[211],"28%":[213],"compared":[214],"state-of-the-art":[216],"velocity-based":[217],"scalers":[218],"workloads,":[221],"consistently":[223],"maintaining":[224],"99%":[225],"SLO":[226],"attainment.":[227]},"counts_by_year":[],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2026-03-25T00:00:00"}
