{"id":"https://openalex.org/W4413378398","doi":"https://doi.org/10.1145/3761813","title":"SAPar: A Surrogate-Assisted DNN Partitioner for Efficient Inferences on Edge TPU Pipelines","display_name":"SAPar: A Surrogate-Assisted DNN Partitioner for Efficient Inferences on Edge TPU Pipelines","publication_year":2025,"publication_date":"2025-08-15","ids":{"openalex":"https://openalex.org/W4413378398","doi":"https://doi.org/10.1145/3761813"},"language":"en","primary_location":{"id":"doi:10.1145/3761813","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3761813","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3761813","source":{"id":"https://openalex.org/S136160450","display_name":"ACM Transactions on Embedded Computing Systems","issn_l":"1539-9087","issn":["1539-9087","1558-3465"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Embedded Computing Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3761813","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5042767630","display_name":"Binqi Sun","orcid":"https://orcid.org/0000-0002-9764-6259"},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Binqi Sun","raw_affiliation_strings":["Technical University of Munich","Technical University of Munich, Munich, Germany"],"raw_orcid":"https://orcid.org/0000-0002-9764-6259","affiliations":[{"raw_affiliation_string":"Technical University of Munich","institution_ids":["https://openalex.org/I62916508"]},{"raw_affiliation_string":"Technical University of Munich, Munich, Germany","institution_ids":["https://openalex.org/I62916508"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bohua Zou","orcid":"https://orcid.org/0009-0000-3228-7702"},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Bohua Zou","raw_affiliation_strings":["Technical University of Munich","Technical University of Munich, Munich Germany"],"raw_orcid":"https://orcid.org/0009-0000-3228-7702","affiliations":[{"raw_affiliation_string":"Technical University of Munich","institution_ids":["https://openalex.org/I62916508"]},{"raw_affiliation_string":"Technical University of Munich, Munich Germany","institution_ids":["https://openalex.org/I62916508"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103165706","display_name":"Yigong Hu","orcid":"https://orcid.org/0009-0000-5320-9439"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yigong Hu","raw_affiliation_strings":["University of Illinois Urbana-Champaign","University of Illinois Urbana-Champaign, Urbana, United States"],"raw_orcid":"https://orcid.org/0009-0000-5320-9439","affiliations":[{"raw_affiliation_string":"University of Illinois Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]},{"raw_affiliation_string":"University of Illinois Urbana-Champaign, Urbana, United States","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078957436","display_name":"Tomasz Kloda","orcid":"https://orcid.org/0000-0003-0822-4976"},"institutions":[{"id":"https://openalex.org/I190497903","display_name":"Laboratoire d'Analyse et d'Architecture des Syst\u00e8mes","ror":"https://ror.org/03vcm6439","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I190497903","https://openalex.org/I193033237","https://openalex.org/I196454796","https://openalex.org/I205747304","https://openalex.org/I4210095849","https://openalex.org/I4210152422","https://openalex.org/I4210159245","https://openalex.org/I4405258862","https://openalex.org/I4405259414"]},{"id":"https://openalex.org/I196454796","display_name":"Institut National des Sciences Appliqu\u00e9es de Toulouse","ror":"https://ror.org/01h8pf755","country_code":"FR","type":"education","lineage":["https://openalex.org/I196454796"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Tomasz Kloda","raw_affiliation_strings":["Institut National des Sciences Appliquees de Toulouse","Laboratoire d'Analyse et d'Architecture des Systemes","Institut National des Sciences Appliquees de Toulouse, Toulouse France","Laboratoire d'Analyse et d'Architecture des Systemes, Toulouse France"],"raw_orcid":"https://orcid.org/0000-0003-0822-4976","affiliations":[{"raw_affiliation_string":"Institut National des Sciences Appliquees de Toulouse","institution_ids":["https://openalex.org/I196454796"]},{"raw_affiliation_string":"Laboratoire d'Analyse et d'Architecture des Systemes","institution_ids":["https://openalex.org/I190497903"]},{"raw_affiliation_string":"Institut National des Sciences Appliquees de Toulouse, Toulouse France","institution_ids":["https://openalex.org/I196454796"]},{"raw_affiliation_string":"Laboratoire d'Analyse et d'Architecture des Systemes, Toulouse France","institution_ids":["https://openalex.org/I190497903"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100398661","display_name":"Ling Wang","orcid":"https://orcid.org/0000-0001-8964-6454"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Wang","raw_affiliation_strings":["Tsinghua University","Tsinghua University, Beijing China"],"raw_orcid":"https://orcid.org/0000-0001-8964-6454","affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Tsinghua University, Beijing China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087114395","display_name":"Tarek Abdelzaher","orcid":"https://orcid.org/0000-0003-3883-7220"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tarek Abdelzaher","raw_affiliation_strings":["University of Illinois Urbana-Champaign","University of Illinois Urbana-Champaign, Urbana, United States"],"raw_orcid":"https://orcid.org/0000-0003-3883-7220","affiliations":[{"raw_affiliation_string":"University of Illinois Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]},{"raw_affiliation_string":"University of Illinois Urbana-Champaign, Urbana, United States","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5060442004","display_name":"Marco Caccamo","orcid":"https://orcid.org/0000-0003-2328-044X"},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Marco Caccamo","raw_affiliation_strings":["Technical University of Munich","Technical University of Munich, Munich Germany"],"raw_orcid":"https://orcid.org/0000-0003-2328-044X","affiliations":[{"raw_affiliation_string":"Technical University of Munich","institution_ids":["https://openalex.org/I62916508"]},{"raw_affiliation_string":"Technical University of Munich, Munich Germany","institution_ids":["https://openalex.org/I62916508"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.5239,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.85257103,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":"24","issue":"5s","first_page":"1","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12707","display_name":"Vehicle License Plate Recognition","score":0.9966999888420105,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12707","display_name":"Vehicle License Plate Recognition","score":0.9966999888420105,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9886999726295471,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9782000184059143,"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/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.6346728801727295},{"id":"https://openalex.org/keywords/pipeline-transport","display_name":"Pipeline transport","score":0.485373854637146},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4376465082168579},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3402901291847229},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.22343480587005615},{"id":"https://openalex.org/keywords/mechanical-engineering","display_name":"Mechanical engineering","score":0.10451292991638184}],"concepts":[{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.6346728801727295},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.485373854637146},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4376465082168579},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3402901291847229},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.22343480587005615},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.10451292991638184}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3761813","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3761813","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3761813","source":{"id":"https://openalex.org/S136160450","display_name":"ACM Transactions on Embedded Computing Systems","issn_l":"1539-9087","issn":["1539-9087","1558-3465"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Embedded Computing Systems","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3761813","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3761813","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3761813","source":{"id":"https://openalex.org/S136160450","display_name":"ACM Transactions on Embedded Computing Systems","issn_l":"1539-9087","issn":["1539-9087","1558-3465"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Embedded Computing Systems","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1034401897","display_name":null,"funder_award_id":"W911NF-17-2-0196","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G1452086972","display_name":null,"funder_award_id":"JUMP 2.0","funder_id":"https://openalex.org/F4320306087","funder_display_name":"Semiconductor Research Corporation"},{"id":"https://openalex.org/G1637379582","display_name":null,"funder_award_id":"W911NF-17-2-0196","funder_id":"https://openalex.org/F4320338295","funder_display_name":"Army Research Laboratory"},{"id":"https://openalex.org/G5259331294","display_name":null,"funder_award_id":"W911NF","funder_id":"https://openalex.org/F4320338295","funder_display_name":"Army Research Laboratory"},{"id":"https://openalex.org/G717040255","display_name":null,"funder_award_id":"CNS 20-38817","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8938883616","display_name":null,"funder_award_id":"20-38817","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320306087","display_name":"Semiconductor Research Corporation","ror":"https://ror.org/047z4n946"},{"id":"https://openalex.org/F4320321114","display_name":"Bundesministerium f\u00fcr Bildung und Forschung","ror":"https://ror.org/04pz7b180"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320337807","display_name":"U.S. Army Combat Capabilities Development Command","ror":null},{"id":"https://openalex.org/F4320338295","display_name":"Army Research Laboratory","ror":"https://ror.org/011hc8f90"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4413378398.pdf","grobid_xml":"https://content.openalex.org/works/W4413378398.grobid-xml"},"referenced_works_count":38,"referenced_works":["https://openalex.org/W398859631","https://openalex.org/W1977655452","https://openalex.org/W2036213524","https://openalex.org/W2063123613","https://openalex.org/W2104984867","https://openalex.org/W2166745914","https://openalex.org/W2399044027","https://openalex.org/W2606958576","https://openalex.org/W2909139149","https://openalex.org/W2912012512","https://openalex.org/W3035595501","https://openalex.org/W3090359607","https://openalex.org/W3097806840","https://openalex.org/W3110923198","https://openalex.org/W3169326577","https://openalex.org/W3176623705","https://openalex.org/W3177525997","https://openalex.org/W3184320749","https://openalex.org/W4205928248","https://openalex.org/W4226422867","https://openalex.org/W4231219113","https://openalex.org/W4238701792","https://openalex.org/W4239308174","https://openalex.org/W4245423414","https://openalex.org/W4293102214","https://openalex.org/W4301045096","https://openalex.org/W4312650633","https://openalex.org/W4312747027","https://openalex.org/W4312936119","https://openalex.org/W4317038519","https://openalex.org/W4322627015","https://openalex.org/W4380366709","https://openalex.org/W4382560998","https://openalex.org/W4390280330","https://openalex.org/W4392601822","https://openalex.org/W4403196715","https://openalex.org/W4403663802","https://openalex.org/W4406414871"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Pipelining":[0],"deep":[1],"neural":[2],"networks":[3],"(DNNs)":[4],"across":[5,58,70,212],"multiple":[6],"Edge":[7,32,49,71,177],"Tensor":[8],"Processing":[9],"Units":[10],"(TPUs)":[11],"can":[12],"enhance":[13],"on-device":[14],"performance":[15,187],"by":[16,81,136,147,208],"increasing":[17],"the":[18,40,52,132,148,205],"capacity":[19],"for":[20,63,120],"DNN":[21,41,76,108,124,215],"parameters":[22],"caching":[23],"and":[24,85,122,142,161,172],"enabling":[25],"pipeline":[26,186],"parallelism.":[27],"Effective":[28],"deployment":[29],"on":[30,175],"pipelined":[31],"TPUs":[33,72],"requires":[34],"a":[35,47,89,105,112,117,219],"partitioning":[36,94,200],"tool":[37],"to":[38,46,93,196,210],"divide":[39],"into":[42],"segments,":[43],"each":[44],"assigned":[45],"different":[48],"TPU":[50,178],"in":[51,199],"pipeline.":[53],"Achieving":[54],"balanced":[55],"workload":[56,68],"distribution":[57],"these":[59],"segments":[60],"is":[61,73,79],"crucial":[62],"optimal":[64],"timing":[65],"performance.":[66],"However,":[67],"balancing":[69],"challenging,":[74],"as":[75],"execution":[77],"time":[78],"influenced":[80],"proprietary":[82],"hardware":[83,179],"architecture":[84],"compiler":[86],"internals,":[87],"forming":[88],"black-box":[90],"function":[91],"inaccessible":[92],"tools.":[95],"To":[96],"address":[97],"this":[98,100],"challenge,":[99],"article":[101],"introduces":[102],"SAPar":[103,182,203],",":[104,168],"new":[106],"surrogate-assisted":[107,118,149],"partitioner":[109,192],"that":[110,181],"integrates":[111],"neighborhood":[113,127,143],"search":[114,128],"engine":[115,129],"with":[116,193,218],"evaluator":[119,152],"effective":[121],"efficient":[123,163],"partitioning.":[125],"The":[126,151],"systematically":[130],"explores":[131],"decision":[133],"space,":[134],"guided":[135],"knowledge":[137],"obtained":[138],"from":[139],"empirical":[140],"insights":[141],"evaluation":[144],"feedback":[145],"provided":[146],"evaluator.":[150],"cooperatively":[153],"applies":[154],"an":[155,162,194],"accurate":[156],"yet":[157],"time-consuming":[158],"latency":[159,207],"profiler":[160],"graph":[164],"transformer-based":[165],"surrogate":[166],"model":[167],"achieving":[169],"both":[170],"precision":[171],"scalability.":[173],"Experiments":[174],"real":[176],"demonstrate":[180],"achieves":[183],"significantly":[184],"better":[185],"than":[188],"Google\u2019s":[189],"current":[190],"profiling-based":[191],"8.82\u00d7":[195],"110\u00d7":[197],"speedup":[198],"time.":[201],"Moreover,":[202],"reduces":[204],"bottleneck":[206],"8.93%":[209],"44.15%":[211],"five":[213],"classic":[214],"models":[216],"compared":[217],"state-of-the-art":[220],"reinforcement":[221],"learning-based":[222],"partitioner.":[223]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-08-22T00:00:00"}