{"id":"https://openalex.org/W7225424338","doi":"https://doi.org/10.5281/zenodo.23997470","title":"RAM-Aware Predictive Computing for Generative AI: Forecasting Future System Memory Demand and Proactively Preparing Host Memory Resources","display_name":"RAM-Aware Predictive Computing for Generative AI: Forecasting Future System Memory Demand and Proactively Preparing Host Memory Resources","publication_year":2026,"publication_date":"2026-11-06","ids":{"openalex":"https://openalex.org/W7225424338","doi":"https://doi.org/10.5281/zenodo.23997470"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.23997470","doi":"https://doi.org/10.5281/zenodo.23997470","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23997470","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.23997470","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100638045","display_name":"Jincheng Zhang","orcid":"https://orcid.org/0000-0001-7332-6704","observed_orcids":["https://orcid.org/0000-0001-7332-6704","https://orcid.org/0009-0005-1860-0009"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Jincheng","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0009-0005-1860-0009","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5100638045"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.6234999895095825,"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":0.6234999895095825,"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":0.2615000009536743,"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/T12127","display_name":"Software System Performance and Reliability","score":0.03280000016093254,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-model-inference","display_name":"large language model inference","score":0.8339999914169312},{"id":"https://openalex.org/keywords/kv-cache-management","display_name":"KV cache management","score":0.8140000104904175},{"id":"https://openalex.org/keywords/memory-fragmentation","display_name":"memory fragmentation","score":0.6269999742507935},{"id":"https://openalex.org/keywords/model-loading","display_name":"model loading","score":0.42899999022483826}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8403000235557556},{"id":"https://openalex.org/C57863822","wikidata":"https://www.wikidata.org/wiki/Q905488","display_name":"Flat memory model","level":4,"score":0.616100013256073},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.5777999758720398},{"id":"https://openalex.org/C63511323","wikidata":"https://www.wikidata.org/wiki/Q908936","display_name":"Interleaved memory","level":4,"score":0.4984999895095825},{"id":"https://openalex.org/C3720319","wikidata":"https://www.wikidata.org/wiki/Q5015937","display_name":"Cache-only memory architecture","level":5,"score":0.47850000858306885},{"id":"https://openalex.org/C171675096","wikidata":"https://www.wikidata.org/wiki/Q1143380","display_name":"Extended memory","level":4,"score":0.4537999927997589},{"id":"https://openalex.org/C30390489","wikidata":"https://www.wikidata.org/wiki/Q4680748","display_name":"Adaptive memory","level":3,"score":0.4535999894142151},{"id":"https://openalex.org/C51290061","wikidata":"https://www.wikidata.org/wiki/Q1936765","display_name":"Uniform memory access","level":4,"score":0.4462999999523163},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.42739999294281006},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4253000020980835},{"id":"https://openalex.org/C74426580","wikidata":"https://www.wikidata.org/wiki/Q719484","display_name":"Memory map","level":3,"score":0.39590001106262207},{"id":"https://openalex.org/C126831891","wikidata":"https://www.wikidata.org/wiki/Q221673","display_name":"Host (biology)","level":2,"score":0.3935999870300293},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.38679999113082886},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.3684000074863434},{"id":"https://openalex.org/C162262903","wikidata":"https://www.wikidata.org/wiki/Q343527","display_name":"Allocator","level":2,"score":0.350600004196167},{"id":"https://openalex.org/C34339311","wikidata":"https://www.wikidata.org/wiki/Q1050390","display_name":"C dynamic memory allocation","level":4,"score":0.3458999991416931},{"id":"https://openalex.org/C92855701","wikidata":"https://www.wikidata.org/wiki/Q5830907","display_name":"Computer memory","level":3,"score":0.3384000062942505},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.33709999918937683},{"id":"https://openalex.org/C2779602883","wikidata":"https://www.wikidata.org/wiki/Q15544750","display_name":"Memory architecture","level":2,"score":0.3237999975681305},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3181000053882599},{"id":"https://openalex.org/C39528615","wikidata":"https://www.wikidata.org/wiki/Q1229610","display_name":"Distributed shared memory","level":5,"score":0.31139999628067017},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C100800780","wikidata":"https://www.wikidata.org/wiki/Q1175867","display_name":"Memory controller","level":3,"score":0.295199990272522},{"id":"https://openalex.org/C82687282","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Auxiliary memory","level":2,"score":0.29269999265670776},{"id":"https://openalex.org/C41036726","wikidata":"https://www.wikidata.org/wiki/Q844824","display_name":"Physical address","level":3,"score":0.29159998893737793},{"id":"https://openalex.org/C2780609101","wikidata":"https://www.wikidata.org/wiki/Q17156588","display_name":"Resource management (computing)","level":2,"score":0.289900004863739},{"id":"https://openalex.org/C118702147","wikidata":"https://www.wikidata.org/wiki/Q189396","display_name":"Dynamic random-access memory","level":3,"score":0.2851000130176544},{"id":"https://openalex.org/C93446704","wikidata":"https://www.wikidata.org/wiki/Q449328","display_name":"Registered memory","level":3,"score":0.27480000257492065},{"id":"https://openalex.org/C152890283","wikidata":"https://www.wikidata.org/wiki/Q4129922","display_name":"Computing with Memory","level":5,"score":0.274399995803833}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.23997470","doi":"https://doi.org/10.5281/zenodo.23997470","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23997470","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},{"id":"doi:10.5281/zenodo.23997471","doi":"https://doi.org/10.5281/zenodo.23997471","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23997471","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"doi:10.5281/zenodo.23997470","doi":"https://doi.org/10.5281/zenodo.23997470","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23997470","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7225392849","https://openalex.org/W7225412337","https://openalex.org/W7225267314","https://openalex.org/W7225375373","https://openalex.org/W7225409378","https://openalex.org/W7225376051","https://openalex.org/W7225423500","https://openalex.org/W7225400169","https://openalex.org/W7225419399","https://openalex.org/W7225432460"],"abstract_inverted_index":{"The":[0,111,255,316],"increasing":[1],"deployment":[2],"of":[3,64,67,93,102,349],"generative":[4,350],"artificial":[5],"intelligence":[6],"(AI)":[7],"systems":[8],"has":[9],"introduced":[10],"substantial":[11],"demands":[12],"on":[13],"host":[14,70,103,231],"system":[15,51,95],"memory,":[16,283],"particularly":[17],"in":[18,53,193],"large":[19],"language":[20],"model":[21,26,36,125,275],"(LLM)":[22],"inference":[23,120,314],"environments":[24],"involving":[25],"loading,":[27],"CPU-based":[28],"preprocessing,":[29],"memory-intensive":[30],"request":[31,122],"handling,":[32],"KV":[33,278],"cache":[34],"management,":[35],"offloading,":[37],"and":[38,49,98,138,157,164,187,222,241,252,264,288,312,339,347],"concurrent":[39,132],"execution.":[40],"Although":[41],"existing":[42],"memory":[43,52,71,104,116,135,149,153,166,170,175,185,211,216,218,239,244,262,268,302,306,324,337],"management":[44,191,342],"techniques":[45],"primarily":[46,173],"allocate,":[47],"reclaim,":[48],"optimize":[50],"response":[54],"to":[55,182,201,229,297,358],"current":[56],"computational":[57,108],"requirements,":[58],"they":[59],"generally":[60],"provide":[61],"limited":[62],"consideration":[63],"the":[65,91,99,266,332,345],"possibility":[66],"anticipating":[68],"future":[69,94,115,184,272],"demand":[72,147],"before":[73,106],"corresponding":[74],"workloads":[75,109],"arrive.":[76],"This":[77],"paper":[78],"introduces":[79],"RAM-Aware":[80],"Predictive":[81],"Computing":[82],"for":[83,208,271],"Generative":[84],"AI":[85,351],"(RAM-PCGA),":[86],"a":[87],"framework":[88,113,256,318],"that":[89,237,322],"investigates":[90,331],"prediction":[92,209],"RAM":[96,146,232],"requirements":[97,186,240,245],"proactive":[100,152,323],"preparation":[101,325],"resources":[105],"anticipated":[107,139],"materialize.":[110],"proposed":[112,317],"estimates":[114],"consumption":[117,150,263],"using":[118],"historical":[119],"workloads,":[121],"arrival":[123],"patterns,":[124,251],"characteristics,":[126],"prompt":[127],"lengths,":[128,131],"expected":[129],"generation":[130],"execution":[133],"levels,":[134],"allocation":[136,156,214,248,304],"histories,":[137],"model-loading":[140,159],"activities.":[141],"Its":[142,353],"architecture":[143],"integrates":[144],"predictive":[145],"modeling,":[148],"decomposition,":[151],"reservation,":[154],"adaptive":[155],"reclamation,":[158],"preparation,":[160],"memory-aware":[161],"workload":[162],"scheduling,":[163],"uncertainty-aware":[165],"protection.":[167],"Unlike":[168],"conventional":[169],"monitoring,":[171],"which":[172,335],"identifies":[174],"pressure":[176],"after":[177],"it":[178,330],"develops,":[179],"RAM-PCGA":[180],"attempts":[181],"estimate":[183],"determine":[188],"appropriate":[189],"resource":[190,341],"actions":[192],"advance.":[194],"A":[195],"unified":[196],"optimization":[197,253],"formulation":[198],"is":[199,227,295,326],"developed":[200],"minimize":[202],"memory-related":[203],"operational":[204],"expenditure":[205],"while":[206],"accounting":[207],"errors,":[210],"reservation":[212],"overhead,":[213,287],"latency,":[215,305],"fragmentation,":[217],"pressure,":[219],"swapping":[220],"costs,":[221],"out-of-memory":[223,310],"risks.":[224],"Particular":[225],"attention":[226],"given":[228],"distinguishing":[230],"from":[233],"GPU":[234,242],"VRAM,":[235],"recognizing":[236],"CPU-side":[238],"device":[243],"exhibit":[246],"different":[247],"mechanisms,":[249],"utilization":[250],"constraints.":[254],"also":[257],"distinguishes":[258],"between":[259],"predicting":[260,265],"aggregate":[261],"individual":[267],"components":[269],"responsible":[270],"demand,":[273],"including":[274],"weights,":[276],"CPU-resident":[277],"caches,":[279],"intermediate":[280],"tensors,":[281],"pinned":[282],"preprocessing":[284],"buffers,":[285],"runtime":[286],"reusable":[289],"data":[290],"structures.":[291],"An":[292],"experimental":[293],"methodology":[294],"established":[296],"evaluate":[298],"forecasting":[299,338],"accuracy,":[300],"peak":[301],"prediction,":[303],"utilization,":[307],"swap":[308],"activity,":[309],"frequency,":[311],"overall":[313],"efficiency.":[315],"does":[319],"not":[320],"assume":[321],"universally":[327],"beneficial;":[328],"rather,":[329],"conditions":[333],"under":[334],"accurate":[336],"anticipatory":[340],"can":[343],"improve":[344],"reliability":[346],"efficiency":[348],"infrastructure.":[352],"practical":[354],"effectiveness":[355],"remains":[356],"subject":[357],"empirical":[359],"validation.":[360]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-11-07T00:00:00"}