{"id":"https://openalex.org/W7126082235","doi":"https://doi.org/10.1109/comcomap68359.2025.11353145","title":"Adaptive GPU Resource Allocation for Multi-Agent Collaborative Reasoning in Serverless Environments","display_name":"Adaptive GPU Resource Allocation for Multi-Agent Collaborative Reasoning in Serverless Environments","publication_year":2025,"publication_date":"2025-12-14","ids":{"openalex":"https://openalex.org/W7126082235","doi":"https://doi.org/10.1109/comcomap68359.2025.11353145"},"language":null,"primary_location":{"id":"doi:10.1109/comcomap68359.2025.11353145","is_oa":false,"landing_page_url":"https://doi.org/10.1109/comcomap68359.2025.11353145","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Computing, Communications and IoT Applications (ComComAp)","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":"https://openalex.org/A5019746763","display_name":"G Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I193531525","display_name":"George Washington University","ror":"https://ror.org/00y4zzh67","country_code":"US","type":"education","lineage":["https://openalex.org/I193531525"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Guilin Zhang","raw_affiliation_strings":["George Washington University,Department of Engineering Management and Systems Engineering,USA"],"affiliations":[{"raw_affiliation_string":"George Washington University,Department of Engineering Management and Systems Engineering,USA","institution_ids":["https://openalex.org/I193531525"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121487655","display_name":"Wulan Guo","orcid":null},"institutions":[{"id":"https://openalex.org/I193531525","display_name":"George Washington University","ror":"https://ror.org/00y4zzh67","country_code":"US","type":"education","lineage":["https://openalex.org/I193531525"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wulan Guo","raw_affiliation_strings":["George Washington University,Department of Engineering Management and Systems Engineering,USA"],"affiliations":[{"raw_affiliation_string":"George Washington University,Department of Engineering Management and Systems Engineering,USA","institution_ids":["https://openalex.org/I193531525"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001214136","display_name":"Ziqi Tan","orcid":"https://orcid.org/0009-0001-8858-2193"},"institutions":[{"id":"https://openalex.org/I193531525","display_name":"George Washington University","ror":"https://ror.org/00y4zzh67","country_code":"US","type":"education","lineage":["https://openalex.org/I193531525"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziqi Tan","raw_affiliation_strings":["George Washington University,Department of Engineering Management and Systems Engineering,USA"],"affiliations":[{"raw_affiliation_string":"George Washington University,Department of Engineering Management and Systems Engineering,USA","institution_ids":["https://openalex.org/I193531525"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5019746763"],"corresponding_institution_ids":["https://openalex.org/I193531525"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.75261479,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"334","last_page":"339"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.09740000218153,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.09740000218153,"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":0.07190000265836716,"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":0.06750000268220901,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.698199987411499},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6662999987602234},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.597599983215332},{"id":"https://openalex.org/keywords/resource-allocation","display_name":"Resource allocation","score":0.5521000027656555},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4973999857902527},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4650000035762787},{"id":"https://openalex.org/keywords/symmetric-multiprocessor-system","display_name":"Symmetric multiprocessor system","score":0.43070000410079956},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.36970001459121704}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.866100013256073},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.698199987411499},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6769000291824341},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6662999987602234},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.597599983215332},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.5521000027656555},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4973999857902527},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4650000035762787},{"id":"https://openalex.org/C172430144","wikidata":"https://www.wikidata.org/wiki/Q17111997","display_name":"Symmetric multiprocessor system","level":2,"score":0.43070000410079956},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.36970001459121704},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.3634999990463257},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.35499998927116394},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.3057999908924103},{"id":"https://openalex.org/C127964446","wikidata":"https://www.wikidata.org/wiki/Q1092142","display_name":"Computational resource","level":3,"score":0.301800012588501},{"id":"https://openalex.org/C5119721","wikidata":"https://www.wikidata.org/wiki/Q220501","display_name":"Quality of service","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C2780609101","wikidata":"https://www.wikidata.org/wiki/Q17156588","display_name":"Resource management (computing)","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.28360000252723694},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.2750000059604645},{"id":"https://openalex.org/C2984822820","wikidata":"https://www.wikidata.org/wiki/Q1123036","display_name":"Processor scheduling","level":3,"score":0.26339998841285706},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.257099986076355}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/comcomap68359.2025.11353145","is_oa":false,"landing_page_url":"https://doi.org/10.1109/comcomap68359.2025.11353145","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Computing, Communications and IoT Applications (ComComAp)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9","score":0.583637535572052}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W4390263770","https://openalex.org/W4408362605","https://openalex.org/W4413320160","https://openalex.org/W7117450920"],"related_works":[],"abstract_inverted_index":{"Multi-agent":[0],"systems":[1,25,176],"powered":[2],"by":[3],"large":[4],"language":[5],"models":[6],"have":[7],"emerged":[8],"as":[9],"a":[10,168],"promising":[11],"paradigm":[12],"for":[13,46,79,171],"solving":[14],"complex":[15],"reasoning":[16],"tasks":[17],"through":[18],"collaborative":[19],"intelligence.":[20],"However,":[21],"efficiently":[22],"deploying":[23,172],"these":[24],"on":[26,89,177],"serverless":[27,134,178],"GPU":[28,54,86,162,179],"platforms":[29],"presents":[30,51],"significant":[31],"resource":[32,55,96],"allocation":[33,56,151],"challenges":[34],"due":[35],"to":[36,64,71],"heterogeneous":[37,113,145],"agent":[38,92],"workloads,":[39],"varying":[40],"computational":[41,114],"demands,":[42],"and":[43,94,119,129,155,161],"the":[44],"need":[45],"cost-effective":[47],"scaling.":[48],"This":[49],"paper":[50],"an":[52,75],"adaptive":[53,150],"framework":[57,107,166],"that":[58,149],"achieves":[59],"85%":[60],"latency":[61],"reduction":[62],"compared":[63],"round-robin":[65,156],"scheduling":[66],"while":[67,101],"maintaining":[68,102],"comparable":[69],"throughput":[70],"static":[72,153],"allocation,":[73],"using":[74],"O(N)":[76],"complexity":[77],"algorithm":[78],"real-time":[80],"adaptation.":[81],"Our":[82],"approach":[83],"dynamically":[84],"allocates":[85],"resources":[87],"based":[88],"workload":[90,124],"characteristics,":[91],"priorities,":[93],"minimum":[95],"requirements,":[97],"enabling":[98],"efficient":[99],"utilization":[100,163],"quality":[103],"of":[104],"service.":[105],"The":[106,165],"addresses":[108],"three":[109],"key":[110],"challenges:":[111],"(1)":[112],"demands":[115],"across":[116,158],"lightweight":[117],"coordinators":[118],"heavyweight":[120],"specialists,":[121],"(2)":[122],"dynamic":[123],"fluctuations":[125],"requiring":[126],"millisecond-scale":[127],"reallocation,":[128],"(3)":[130],"capacity":[131],"constraints":[132],"in":[133],"environments.":[135],"Through":[136],"comprehensive":[137],"simulations":[138],"modeling":[139],"realistic":[140],"multi-agent":[141],"workflows":[142],"with":[143],"four":[144],"agents,":[146],"we":[147],"demonstrate":[148],"outperforms":[152],"equal":[154],"strategies":[157],"latency,":[159],"cost,":[160],"metrics.":[164],"provides":[167],"practical":[169],"solution":[170],"cost-efficient":[173],"multiagent":[174],"AI":[175],"infrastructure.":[180]},"counts_by_year":[],"updated_date":"2026-02-01T03:34:12.195049","created_date":"2026-01-30T00:00:00"}