{"id":"https://openalex.org/W7154269314","doi":"https://doi.org/10.1145/3802513.3803483","title":"ConfigSpec: Profiling-Based Configuration Selection for Distributed Edge-Cloud Speculative LLM Serving","display_name":"ConfigSpec: Profiling-Based Configuration Selection for Distributed Edge-Cloud Speculative LLM Serving","publication_year":2026,"publication_date":"2026-04-14","ids":{"openalex":"https://openalex.org/W7154269314","doi":"https://doi.org/10.1145/3802513.3803483"},"language":null,"primary_location":{"id":"doi:10.1145/3802513.3803483","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3802513.3803483","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 4th International Workshop on Testing Distributed Internet of Things Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2604.09722","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069413313","display_name":"Xiangchen Li","orcid":"https://orcid.org/0000-0002-5293-5348"},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiangchen Li","raw_affiliation_strings":["Virginia Tech, Blacksburg, Virginia, USA"],"raw_orcid":"https://orcid.org/0000-0002-5293-5348","affiliations":[{"raw_affiliation_string":"Virginia Tech, Blacksburg, Virginia, USA","institution_ids":["https://openalex.org/I859038795"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019689955","display_name":"Saeid Ghafouri","orcid":"https://orcid.org/0000-0003-3799-5702"},"institutions":[{"id":"https://openalex.org/I126231945","display_name":"Queen's University Belfast","ror":"https://ror.org/00hswnk62","country_code":"GB","type":"education","lineage":["https://openalex.org/I126231945"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Saeid Ghafouri","raw_affiliation_strings":["Queen's University Belfast, Belfast, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0003-3799-5702","affiliations":[{"raw_affiliation_string":"Queen's University Belfast, Belfast, United Kingdom","institution_ids":["https://openalex.org/I126231945"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125658018","display_name":"Jiakun Fan","orcid":null},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiakun Fan","raw_affiliation_strings":["Virginia Tech, Blacksburg, Virginia, USA"],"raw_orcid":"https://orcid.org/0009-0004-4563-8697","affiliations":[{"raw_affiliation_string":"Virginia Tech, Blacksburg, Virginia, USA","institution_ids":["https://openalex.org/I859038795"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038372328","display_name":"Babar Ali","orcid":"https://orcid.org/0000-0003-0542-848X"},"institutions":[{"id":"https://openalex.org/I126231945","display_name":"Queen's University Belfast","ror":"https://ror.org/00hswnk62","country_code":"GB","type":"education","lineage":["https://openalex.org/I126231945"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Babar Ali","raw_affiliation_strings":["Queen's University Belfast, Belfast, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0003-0542-848X","affiliations":[{"raw_affiliation_string":"Queen's University Belfast, Belfast, United Kingdom","institution_ids":["https://openalex.org/I126231945"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086758775","display_name":"Hans Vandierendonck","orcid":"https://orcid.org/0000-0001-5868-9259"},"institutions":[{"id":"https://openalex.org/I126231945","display_name":"Queen's University Belfast","ror":"https://ror.org/00hswnk62","country_code":"GB","type":"education","lineage":["https://openalex.org/I126231945"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Hans Vandierendonck","raw_affiliation_strings":["Queen's University Belfast, Belfast, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0001-5868-9259","affiliations":[{"raw_affiliation_string":"Queen's University Belfast, Belfast, United Kingdom","institution_ids":["https://openalex.org/I126231945"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5005410613","display_name":"Dimitrios S. Nikolopoulos","orcid":"https://orcid.org/0000-0003-0217-8307"},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dimitrios S. Nikolopoulos","raw_affiliation_strings":["Virginia Tech, Blacksburg, Virginia, USA"],"raw_orcid":"https://orcid.org/0000-0003-0217-8307","affiliations":[{"raw_affiliation_string":"Virginia Tech, Blacksburg, Virginia, USA","institution_ids":["https://openalex.org/I859038795"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.45588578,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.6288999915122986,"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/T12127","display_name":"Software System Performance and Reliability","score":0.6288999915122986,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.09239999949932098,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.07500000298023224,"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/selection","display_name":"Selection (genetic algorithm)","score":0.47189998626708984},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.29510000348091125},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.27399998903274536},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.2612000107765198},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.24709999561309814}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6330999732017517},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.47189998626708984},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.29510000348091125},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.27399998903274536},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2619999945163727},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.2551000118255615},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.24709999561309814},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.24140000343322754},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.23970000445842743}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3802513.3803483","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3802513.3803483","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 4th International Workshop on Testing Distributed Internet of Things Systems","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2604.09722","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2604.09722","pdf_url":"https://arxiv.org/pdf/2604.09722","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2604.09722","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2604.09722","pdf_url":"https://arxiv.org/pdf/2604.09722","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":[],"awards":[{"id":"https://openalex.org/G3294571139","display_name":"U.S.-Ireland R&D Partnership:CNS:Small:SWEET: Hardware and Software for Sustainable Wearable Edge Intelligence","funder_award_id":"2315851","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7111886443","display_name":null,"funder_award_id":"USI-226","funder_id":"https://openalex.org/F4320316531","funder_display_name":"Department for the Economy"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320316531","display_name":"Department for the Economy","ror":"https://ror.org/0161w0r98"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W4404134026","https://openalex.org/W4405717632","https://openalex.org/W4411725577","https://openalex.org/W4412945475","https://openalex.org/W4416955988"],"related_works":[],"abstract_inverted_index":{"Speculative":[0],"decoding":[1],"enables":[2],"collaborative":[3],"Large":[4],"Language":[5],"Model":[6],"(LLM)":[7],"inference":[8],"across":[9,85,92],"cloud":[10],"and":[11,26,45,65,68,74,82,96,122,144],"edge":[12,47,63,94],"by":[13,108],"separating":[14],"lightweight":[15],"token":[16],"drafting":[17,70],"from":[18],"heavyweight":[19],"verification.":[20],"While":[21],"prior":[22],"systems":[23],"show":[24],"performance":[25],"cost":[27,80,121,134],"benefits,":[28],"practical":[29],"deployment":[30],"requires":[31],"navigating":[32],"a":[33,53,130],"large":[34],"configuration":[35,88,161,172],"space":[36],"spanning":[37],"draft":[38,112],"model":[39,113],"variants,":[40],"quantisation":[41],"levels,":[42],"speculative":[43,58,116],"lengths,":[44],"heterogeneous":[46],"devices.":[48],"This":[49],"paper":[50],"presents":[51],"ConfigSpec,":[52],"configurationselection":[54],"framework":[55],"for":[56,139,149,170],"distributed":[57],"LLM":[59,98,177],"serving.":[60],"ConfigSpec":[61],"profiles":[62],"devices":[64],"draft-target":[66],"alignment,":[67],"models":[69],"throughput,":[71],"acceptance":[72,142],"rate,":[73],"power":[75,152],"to":[76,126,129],"evaluate":[77],"goodput,":[78],"verification":[79],"efficiency,":[81],"energy":[83,123,145],"efficiency":[84,124],"the":[86,109,136,147,168],"joint":[87],"space.":[89],"Our":[90],"analysis":[91],"three":[93],"platforms":[95],"two":[97],"families":[99],"reveals":[100],"structurally":[101],"conflicting":[102],"optima.":[103],"Firstly,":[104],"goodput":[105],"is":[106],"maximised":[107],"smallest,":[110],"fastest":[111],"at":[114],"device-dependent":[115],"lengths":[117],"(K*=2-10).":[118],"Secondly,":[119],"both":[120],"converge":[125],"K=2":[127],"due":[128],"dominant":[131],"bonus-token":[132],"effect-with":[133],"favouring":[135,146],"largest":[137],"drafter":[138],"its":[140,150],"high":[141],"rate":[143],"smallest":[148],"low":[151],"draw.":[153],"These":[154],"conflicts":[155],"confirm":[156],"that":[157],"no":[158],"single":[159],"fixed":[160],"can":[162],"simultaneously":[163],"optimise":[164],"all":[165],"objectives,":[166],"underscoring":[167],"need":[169],"profiling-based":[171],"selection":[173],"in":[174],"disaggregated":[175],"edge-cloud":[176],"inference.":[178]},"counts_by_year":[],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2026-04-15T00:00:00"}
