{"id":"https://openalex.org/W3049586678","doi":"https://doi.org/10.1109/tc.2021.3066883","title":"DORY: Automatic End-to-End Deployment of Real-World DNNs on Low-Cost IoT MCUs","display_name":"DORY: Automatic End-to-End Deployment of Real-World DNNs on Low-Cost IoT MCUs","publication_year":2021,"publication_date":"2021-03-18","ids":{"openalex":"https://openalex.org/W3049586678","doi":"https://doi.org/10.1109/tc.2021.3066883","mag":"3049586678"},"language":"en","primary_location":{"id":"doi:10.1109/tc.2021.3066883","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2021.3066883","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2008.07127","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Alessio Burrello","orcid":"https://orcid.org/0000-0002-6215-8220"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Alessio Burrello","raw_affiliation_strings":["Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Angelo Garofalo","orcid":"https://orcid.org/0000-0002-7495-6895"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Angelo Garofalo","raw_affiliation_strings":["Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Nazareno Bruschi","orcid":null},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Nazareno Bruschi","raw_affiliation_strings":["Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Giuseppe Tagliavini","orcid":"https://orcid.org/0000-0002-9221-4633"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Giuseppe Tagliavini","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Davide Rossi","orcid":"https://orcid.org/0000-0002-0651-5393"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Davide Rossi","raw_affiliation_strings":["Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"last","author":{"id":null,"display_name":"Francesco Conti","orcid":"https://orcid.org/0000-0002-7924-933X"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Francesco Conti","raw_affiliation_strings":["Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I9360294"],"apc_list":null,"apc_paid":null,"fwci":19.5719,"has_fulltext":false,"cited_by_count":105,"citation_normalized_percentile":{"value":0.99800736,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":100},"biblio":{"volume":"70","issue":"8","first_page":"1253","last_page":"1268"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.5141000151634216,"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":0.5141000151634216,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.1679999977350235,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.04479999840259552,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/microcontroller","display_name":"Microcontroller","score":0.5156999826431274},{"id":"https://openalex.org/keywords/static-random-access-memory","display_name":"Static random-access memory","score":0.4848000109195709},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.4641999900341034},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4350999891757965},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.43320000171661377},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.42739999294281006},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.41440001130104065},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4104999899864197},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.39719998836517334}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8033000230789185},{"id":"https://openalex.org/C173018170","wikidata":"https://www.wikidata.org/wiki/Q165678","display_name":"Microcontroller","level":2,"score":0.5156999826431274},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5078999996185303},{"id":"https://openalex.org/C68043766","wikidata":"https://www.wikidata.org/wiki/Q267416","display_name":"Static random-access memory","level":2,"score":0.4848000109195709},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.4641999900341034},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4350999891757965},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.43320000171661377},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.42739999294281006},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.41440001130104065},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4104999899864197},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.39719998836517334},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.38670000433921814},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3849000036716461},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3747999966144562},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.36820000410079956},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.3598000109195709},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3587000072002411},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3422999978065491},{"id":"https://openalex.org/C81860439","wikidata":"https://www.wikidata.org/wiki/Q251212","display_name":"Internet of Things","level":2,"score":0.3303000032901764},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3068000078201294},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.30630001425743103},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.30329999327659607},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.29670000076293945},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.28760001063346863},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C157170001","wikidata":"https://www.wikidata.org/wiki/Q4781507","display_name":"Applications of artificial intelligence","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C24590314","wikidata":"https://www.wikidata.org/wiki/Q336038","display_name":"Wireless sensor network","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tc.2021.3066883","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2021.3066883","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2008.07127","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2008.07127","pdf_url":"https://arxiv.org/pdf/2008.07127","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:cris.unibo.it:11585/846999","is_oa":true,"landing_page_url":"http://hdl.handle.net/11585/846999","pdf_url":null,"source":{"id":"https://openalex.org/S4306402579","display_name":"Archivio istituzionale della ricerca (Alma Mater Studiorum Universit\u00e0 di Bologna)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210117483","host_organization_name":"Istituto di Ematologia di Bologna","host_organization_lineage":["https://openalex.org/I4210117483"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2008.07127","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2008.07127","pdf_url":"https://arxiv.org/pdf/2008.07127","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7774680702","display_name":null,"funder_award_id":"826060","funder_id":"https://openalex.org/F4320327207","funder_display_name":"Electronic Components and Systems for European Leadership"}],"funders":[{"id":"https://openalex.org/F4320327207","display_name":"Electronic Components and Systems for European Leadership","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W1970825693","https://openalex.org/W2008314574","https://openalex.org/W2117696986","https://openalex.org/W2132829148","https://openalex.org/W2169500297","https://openalex.org/W2233304223","https://openalex.org/W2442974303","https://openalex.org/W2563468918","https://openalex.org/W2585140699","https://openalex.org/W2586992378","https://openalex.org/W2594836184","https://openalex.org/W2762644836","https://openalex.org/W2806576037","https://openalex.org/W2889339831","https://openalex.org/W2891465615","https://openalex.org/W2899625887","https://openalex.org/W2909953480","https://openalex.org/W2940862705","https://openalex.org/W2946485102","https://openalex.org/W2963163009","https://openalex.org/W2963255460","https://openalex.org/W2970796375","https://openalex.org/W2979644612","https://openalex.org/W2995086119","https://openalex.org/W2999093589","https://openalex.org/W3013186616","https://openalex.org/W3017521908","https://openalex.org/W3033079779","https://openalex.org/W4288083528","https://openalex.org/W6645576446","https://openalex.org/W6677580257","https://openalex.org/W6737664043","https://openalex.org/W6746312620","https://openalex.org/W6747876305","https://openalex.org/W6748224102","https://openalex.org/W6762718338","https://openalex.org/W6775588655","https://openalex.org/W6779527419","https://openalex.org/W6780827055","https://openalex.org/W6784225549","https://openalex.org/W6797015341"],"related_works":[],"abstract_inverted_index":{"The":[0],"deployment":[1],"of":[2,13,56,98,167,217],"Deep":[3,23],"Neural":[4],"Networks":[5],"(DNNs)":[6],"on":[7,63,89,176,202,206,225],"end-nodes":[8,28],"at":[9],"the":[10,14,57,117,146,168,177,190,199,242,245,250],"extreme":[11],"edge":[12],"Internet-of-Things":[15],"is":[16],"a":[17,106,156,218],"critical":[18],"enabler":[19],"to":[20,39,80,86,131,141,185],"support":[21],"pervasive":[22],"Learning-enhanced":[24],"applications.":[25],"Low-Cost":[26],"MCU-based":[27],"have":[29],"limited":[30],"on-chip":[31,99,135],"memory":[32,51,58,114],"and":[33,43,70,134,137,195,249],"often":[34],"replace":[35],"caches":[36],"with":[37,93,149],"scratchpads,":[38],"reduce":[40],"area":[41],"overheads":[42],"increase":[44],"energy":[45],"efficiency":[46],"-":[47,82,230,241,253],"requiring":[48],"explicit":[49],"DMA-based":[50],"transfers":[52,136],"between":[53],"different":[54],"levels":[55],"hierarchy.":[59],"Mapping":[60],"modern":[61],"DNNs":[62,88],"these":[64],"systems":[65],"requires":[66],"aggressive":[67],"topology-dependent":[68],"tiling":[69,104],"double-buffering.":[71],"In":[72],"this":[73,180],"work,":[74],"we":[75,161],"propose":[76],"DORY":[77,102,144,182,243],"(Deployment":[78],"Oriented":[79],"memoRY)":[81],"an":[83,203,234],"automatic":[84],"tool":[85],"deploy":[87],"low":[90],"cost":[91],"MCUs":[92],"typically":[94],"less":[95],"than":[96,189,198,233],"1MB":[97],"SRAM":[100],"memory.":[101],"abstracts":[103],"as":[105,254],"Constraint":[107],"Programming":[108],"(CP)":[109],"problem:":[110],"it":[111,126],"maximizes":[112],"L1":[113],"utilization":[115],"under":[116],"topological":[118],"constraints":[119],"imposed":[120],"by":[121],"each":[122],"DNN":[123],"layer.":[124],"Then,":[125],"generates":[127],"ANSI":[128],"C":[129],"code":[130],"orchestrate":[132],"off-":[133],"computation":[138],"phases.":[139],"Furthermore,":[140],"maximize":[142],"speed,":[143],"augments":[145],"CP":[147],"formulation":[148],"heuristics":[150,252],"promoting":[151],"performance-effective":[152],"tile":[153],"sizes.":[154],"As":[155],"case":[157],"study":[158],"for":[159],"DORY,":[160],"target":[162],"GreenWaves":[163,191],"Technologies":[164],"GAP8,":[165],"one":[166],"most":[169],"advanced":[170],"parallel":[171],"ultra-low":[172],"power":[173],"MCU-class":[174],"devices":[175],"market.":[178],"On":[179],"device,":[181],"achieves":[183],"up":[184],"2.5\u00d7":[186],"better":[187,197,232],"MAC/cycle":[188],"proprietary":[192],"software":[193],"solution":[194],"18.1\u00d7":[196],"state-of-the-art":[200],"result":[201],"STM32-H743":[204],"MCU":[205],"single":[207],"layers.":[208],"Using":[209],"our":[210,239],"tool,":[211],"GAP-8":[212],"can":[213],"perform":[214],"end-to-end":[215],"inference":[216],"1.0-MobileNet-128":[219],"network":[220],"consuming":[221],"just":[222],"63":[223],"pJ/MAC":[224],"average":[226],"@":[227],"4.3":[228],"fps":[229],"15.4\u00d7":[231],"STM32-H743.":[235],"We":[236],"release":[237],"all":[238],"developments":[240],"framework,":[244],"optimized":[246],"backend":[247],"kernels,":[248],"related":[251],"open-source":[255],"software.":[256]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":20},{"year":2024,"cited_by_count":30},{"year":2023,"cited_by_count":27},{"year":2022,"cited_by_count":15},{"year":2021,"cited_by_count":11}],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2020-08-21T00:00:00"}