{"id":"https://openalex.org/W4399323669","doi":"https://doi.org/10.1145/3643832.3661878","title":"Pantheon: Preemptible Multi-DNN Inference on Mobile Edge GPUs","display_name":"Pantheon: Preemptible Multi-DNN Inference on Mobile Edge GPUs","publication_year":2024,"publication_date":"2024-06-03","ids":{"openalex":"https://openalex.org/W4399323669","doi":"https://doi.org/10.1145/3643832.3661878"},"language":"en","primary_location":{"id":"doi:10.1145/3643832.3661878","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3643832.3661878","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3643832.3661878","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3643832.3661878","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003963473","display_name":"L D Han","orcid":"https://orcid.org/0000-0001-5350-6046"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Lixiang Han","raw_affiliation_strings":["City University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":"https://orcid.org/0000-0001-5350-6046","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011140675","display_name":"Zimu Zhou","orcid":"https://orcid.org/0000-0002-5457-6967"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zimu Zhou","raw_affiliation_strings":["City University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":"https://orcid.org/0000-0002-5457-6967","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100419083","display_name":"Zhenjiang Li","orcid":"https://orcid.org/0000-0002-3296-3392"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhenjiang Li","raw_affiliation_strings":["City University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":"https://orcid.org/0000-0002-3296-3392","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I168719708"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5003963473"],"corresponding_institution_ids":["https://openalex.org/I168719708"],"apc_list":null,"apc_paid":null,"fwci":4.5908,"has_fulltext":false,"cited_by_count":21,"citation_normalized_percentile":{"value":0.95090102,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"465","last_page":"478"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9986000061035156,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9986000061035156,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9983000159263611,"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"}},{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9968000054359436,"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/computer-science","display_name":"Computer science","score":0.7587317228317261},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6759664416313171},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5864453315734863},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.3560332953929901},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3537135422229767}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7587317228317261},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6759664416313171},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5864453315734863},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.3560332953929901},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3537135422229767}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3643832.3661878","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3643832.3661878","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3643832.3661878","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3643832.3661878","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3643832.3661878","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3643832.3661878","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7422389112","display_name":null,"funder_award_id":"APRC (9610633)","funder_id":"https://openalex.org/F4320309893","funder_display_name":"City University of Hong Kong"}],"funders":[{"id":"https://openalex.org/F4320309893","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4399323669.pdf"},"referenced_works_count":47,"referenced_works":["https://openalex.org/W1965804146","https://openalex.org/W2002427601","https://openalex.org/W2041616772","https://openalex.org/W2044561013","https://openalex.org/W2097117768","https://openalex.org/W2107034620","https://openalex.org/W2166440675","https://openalex.org/W2194775991","https://openalex.org/W2565600385","https://openalex.org/W2604319603","https://openalex.org/W2767421475","https://openalex.org/W2786171709","https://openalex.org/W2791175987","https://openalex.org/W2792243241","https://openalex.org/W2796438033","https://openalex.org/W2860338957","https://openalex.org/W2920031528","https://openalex.org/W2962677625","https://openalex.org/W2965289829","https://openalex.org/W2970971581","https://openalex.org/W2984200518","https://openalex.org/W3014810041","https://openalex.org/W3046329988","https://openalex.org/W3047147270","https://openalex.org/W3047681172","https://openalex.org/W3110875274","https://openalex.org/W3121074724","https://openalex.org/W3144271226","https://openalex.org/W3183898570","https://openalex.org/W3191321386","https://openalex.org/W3205504880","https://openalex.org/W3211149853","https://openalex.org/W3215253865","https://openalex.org/W4236853429","https://openalex.org/W4306178486","https://openalex.org/W4306250089","https://openalex.org/W4308089797","https://openalex.org/W4317927946","https://openalex.org/W4317927968","https://openalex.org/W4321853806","https://openalex.org/W4380925920","https://openalex.org/W4383704293","https://openalex.org/W4395668837","https://openalex.org/W6887863620","https://openalex.org/W6929237974","https://openalex.org/W6948266939","https://openalex.org/W7004919675"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2382290278","https://openalex.org/W4395014643"],"abstract_inverted_index":{"GPUs":[0,113,145],"are":[1,109,190],"increasingly":[2],"utilized":[3],"for":[4,54,156,211],"running":[5],"DNN":[6,28,212],"tasks":[7,88,170],"on":[8,114,140,150,215],"emerging":[9],"mobile":[10,48,97,116,151],"edge":[11,98],"devices.":[12],"Beyond":[13],"accelerating":[14],"single":[15],"task":[16,55,75,136],"inference,":[17],"their":[18],"value":[19],"is":[20,39,93,181],"also":[21],"particularly":[22],"apparent":[23],"in":[24,36,96,130,224],"efficiently":[25],"executing":[26],"multiple":[27],"tasks,":[29],"which":[30],"often":[31],"have":[32],"strict":[33],"latency":[34],"requirements":[35],"applications.":[37,99],"Preemption":[38],"the":[40,81,183,207],"main":[41],"technology":[42],"to":[43,76,101,164,171],"ensure":[44],"multitasking":[45],"timeliness,":[46],"but":[47,91,138],"edges":[49],"primarily":[50,118],"offer":[51,165],"two":[52],"priorities":[53],"queues,":[56],"and":[57,70,122,175,203,228],"existing":[58],"methods":[59],"thus":[60],"achieve":[61],"only":[62],"coarse-grained":[63],"preemption":[64,125,193],"by":[65,200],"categorizing":[66],"DNNs":[67,153,220],"into":[68],"real-time":[69,74,87,169],"best-effort,":[71],"permitting":[72],"a":[73,216],"preempt":[77,172],"best-effort":[78,176],"ones.":[79],"However,":[80],"efficacy":[82],"diminishes":[83],"significantly":[84],"when":[85],"other":[86,107,174],"run":[89],"concurrently,":[90],"this":[92],"already":[94],"common":[95],"Due":[100],"different":[102],"hardware":[103],"characteristics,":[104],"solutions":[105],"from":[106],"platforms":[108],"unsuitable.":[110],"For":[111],"instance,":[112],"traditional":[115],"devices":[117],"assist":[119],"CPU":[120],"processing":[121],"lack":[123],"special":[124],"support,":[126],"mainly":[127,154],"following":[128],"FIFO":[129],"GPU":[131,185],"scheduling.":[132],"Clouds":[133],"handle":[134],"concurrent":[135],"execution,":[137],"focus":[139],"allocating":[141],"one":[142,157],"or":[143],"more":[144],"per":[146],"complex":[147],"model,":[148],"whereas":[149],"edges,":[152],"vie":[155],"GPU.":[158],"This":[159],"paper":[160],"introduces":[161],"Pantheon,":[162],"designed":[163],"fine-grained":[166],"preemption,":[167],"enabling":[168],"each":[173],"tasks.":[177],"Our":[178],"key":[179],"observation":[180],"that":[182],"two-tier":[184],"stream":[186],"priorities,":[187],"while":[188],"underexplored,":[189],"sufficient.":[191],"Efficient":[192],"can":[194],"be":[195],"realized":[196],"through":[197],"software":[198],"design":[199],"innovative":[201],"scheduling":[202],"novel":[204],"exploitation":[205],"of":[206,219,230],"nested":[208],"redundancy":[209],"principle":[210],"models.":[213],"Evaluation":[214],"diverse":[217],"set":[218],"shows":[221],"substantial":[222],"improvements":[223],"deadline":[225],"miss":[226],"rate":[227],"accuracy":[229],"Pantheon":[231],"over":[232],"state-of-the-art":[233],"methods.":[234]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":15},{"year":2024,"cited_by_count":2}],"updated_date":"2026-06-03T09:05:47.796612","created_date":"2025-10-10T00:00:00"}