{"id":"https://openalex.org/W7131131273","doi":"https://doi.org/10.1109/tpds.2026.3667207","title":"Adaptive Block-Wise Mapping With Intra-Block Resource Allocation for Multi-DNN Workloads on Heterogeneous Accelerator Systems","display_name":"Adaptive Block-Wise Mapping With Intra-Block Resource Allocation for Multi-DNN Workloads on Heterogeneous Accelerator Systems","publication_year":2026,"publication_date":"2026-02-23","ids":{"openalex":"https://openalex.org/W7131131273","doi":"https://doi.org/10.1109/tpds.2026.3667207"},"language":null,"primary_location":{"id":"doi:10.1109/tpds.2026.3667207","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2026.3667207","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","raw_type":"journal-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/A5054162159","display_name":"Zhenyu Nie","orcid":"https://orcid.org/0000-0002-7714-7903"},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhenyu Nie","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0002-7714-7903","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126607489","display_name":"Haotian Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haotian Wang","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0002-0086-6301","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Anthony Theodore Chronopoulos","orcid":"https://orcid.org/0000-0002-0094-1017"},"institutions":[{"id":"https://openalex.org/I4403928334","display_name":"Shanghai Institute for Mathematics and Interdisciplinary Sciences","ror":"https://ror.org/0489cd473","country_code":null,"type":"government","lineage":["https://openalex.org/I4403928334"]}],"countries":[],"is_corresponding":false,"raw_author_name":"Anthony Theodore Chronopoulos","raw_affiliation_strings":["Shanghai Institute for Mathematics and Interdisciplinary Sciences (SIMIS), Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-0094-1017","affiliations":[{"raw_affiliation_string":"Shanghai Institute for Mathematics and Interdisciplinary Sciences (SIMIS), Shanghai, China","institution_ids":["https://openalex.org/I4403928334"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhuo Tang","orcid":"https://orcid.org/0000-0001-9081-8153"},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuo Tang","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0001-9081-8153","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126600434","display_name":"Kenli Li","orcid":null},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kenli Li","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0002-2635-7716","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006345319","display_name":"Chubo Liu","orcid":"https://orcid.org/0000-0002-2372-6715"},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chubo Liu","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0002-2372-6715","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5126633033","display_name":"Zheng Xiao","orcid":null},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Xiao","raw_affiliation_strings":["College of Computer Science and Electronic Engineering, Hunan University, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0003-1144-7599","affiliations":[{"raw_affiliation_string":"College of Computer Science and Electronic Engineering, Hunan University, Hunan, China","institution_ids":["https://openalex.org/I16609230"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.174946,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"37","issue":"4","first_page":"1015","last_page":"1031"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.12549999356269836,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.12549999356269836,"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/T13553","display_name":"Age of Information Optimization","score":0.09210000187158585,"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":0.06069999933242798,"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/computation","display_name":"Computation","score":0.6574000120162964},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.6470999717712402},{"id":"https://openalex.org/keywords/resource-allocation","display_name":"Resource allocation","score":0.6222000122070312},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.580299973487854},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.5630000233650208},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5273000001907349},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.46889999508857727},{"id":"https://openalex.org/keywords/edge-computing","display_name":"Edge computing","score":0.43689998984336853}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8849999904632568},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.7265999913215637},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6574000120162964},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.6470999717712402},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.6222000122070312},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.580299973487854},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5630000233650208},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5273000001907349},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.46889999508857727},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.43689998984336853},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.42100000381469727},{"id":"https://openalex.org/C190793597","wikidata":"https://www.wikidata.org/wiki/Q189768","display_name":"Application layer","level":3,"score":0.3709999918937683},{"id":"https://openalex.org/C2780609101","wikidata":"https://www.wikidata.org/wiki/Q17156588","display_name":"Resource management (computing)","level":2,"score":0.35850000381469727},{"id":"https://openalex.org/C101765175","wikidata":"https://www.wikidata.org/wiki/Q577764","display_name":"Communications system","level":2,"score":0.35740000009536743},{"id":"https://openalex.org/C158207573","wikidata":"https://www.wikidata.org/wiki/Q5747224","display_name":"Heterogeneous network","level":4,"score":0.3472000062465668},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.31779998540878296},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.313400000333786},{"id":"https://openalex.org/C192126672","wikidata":"https://www.wikidata.org/wiki/Q1068715","display_name":"Telecommunications network","level":2,"score":0.289900004863739},{"id":"https://openalex.org/C51332947","wikidata":"https://www.wikidata.org/wiki/Q1172305","display_name":"Shared resource","level":2,"score":0.28040000796318054},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.2721000015735626},{"id":"https://openalex.org/C117468950","wikidata":"https://www.wikidata.org/wiki/Q208074","display_name":"Network layer","level":3,"score":0.25130000710487366},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2026.3667207","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2026.3667207","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W1901129140","https://openalex.org/W1980465669","https://openalex.org/W2067523571","https://openalex.org/W2162232298","https://openalex.org/W2194775991","https://openalex.org/W2442974303","https://openalex.org/W2835955725","https://openalex.org/W2883929540","https://openalex.org/W2962799344","https://openalex.org/W2963163009","https://openalex.org/W2964245886","https://openalex.org/W2969289739","https://openalex.org/W2997106510","https://openalex.org/W3017521908","https://openalex.org/W3025152414","https://openalex.org/W3090495278","https://openalex.org/W3101026687","https://openalex.org/W3110777925","https://openalex.org/W3158233068","https://openalex.org/W3165926952","https://openalex.org/W3176063860","https://openalex.org/W3211853974","https://openalex.org/W4246193833","https://openalex.org/W4249932213","https://openalex.org/W4280489237","https://openalex.org/W4285173593","https://openalex.org/W4293025058","https://openalex.org/W4296473615","https://openalex.org/W4379115905","https://openalex.org/W4385245566","https://openalex.org/W4386764003","https://openalex.org/W4386764246","https://openalex.org/W4387319124","https://openalex.org/W4391164333","https://openalex.org/W4394871723","https://openalex.org/W4395680460","https://openalex.org/W4401507617","https://openalex.org/W4404057417","https://openalex.org/W4404102513","https://openalex.org/W4404955773","https://openalex.org/W4410582562"],"related_works":[],"abstract_inverted_index":{"Deep":[0],"neural":[1],"networks":[2],"(DNNs)":[3],"dominate":[4],"workloads":[5,74,258],"on":[6,43,75,116,179],"cloud":[7],"and":[8,59,90,99,184,207,228,244,259,279],"edge":[9],"platforms.":[10],"Meanwhile,":[11],"the":[12,16,30,44,49,65,85,135,162,180,187,191,199,248,251],"hardware":[13],"platform":[14],"towards":[15],"heterogeneous":[17,76],"system":[18,260],"with":[19,241],"various":[20,257],"accelerators.":[21],"By":[22],"mapping":[23,40,102,158,227],"layers":[24,42,82,96],"to":[25,61,80,83,103,121,174,190,216,273,285],"their":[26],"different":[27],"preferred":[28],"accelerators,":[29],"computation":[31,89,143,203],"cost":[32,144,164],"of":[33,72,137,145,182,202],"each":[34,214],"layer":[35,176,224],"can":[36,47],"be":[37],"reduced.":[38],"While":[39],"these":[41],"same":[45],"accelerator":[46],"reduce":[48,104],"inter-accelerator":[50],"communication":[51,91,106,125,208],"cost.":[52],"These":[53],"two":[54],"costs":[55],"are":[56],"often":[57],"competing":[58],"difficult":[60],"optimize":[62,223],"simultaneously.":[63],"Therefore,":[64],"core":[66],"challenge":[67],"in":[68,276],"achieving":[69,271],"efficient":[70],"execution":[71,277],"DNN":[73],"systems":[77],"is:":[78],"how":[79],"map":[81],"achieve":[84],"best":[86],"trade-off":[87,192],"between":[88],"costs.":[92],"Existing":[93],"works":[94],"group":[95],"into":[97],"blocks":[98,127],"perform":[100],"blockwise":[101],"inter-layer":[105,124],"within":[107,126,213],"blocks.":[108],"However,":[109],"when":[110],"grouping":[111,172,177,188],"layers,":[112],"they":[113],"typically":[114],"rely":[115],"modelagnostic":[117],"rules,":[118],"which":[119],"fail":[120],"hide":[122],"critical":[123],"for":[128,160],"diverse":[129],"DNNs.":[130],"Moreover,":[131],"after":[132],"block":[133,215],"mapping,":[134],"lack":[136],"intra-block":[138,229],"resource":[139,219,230],"allocation":[140,201],"further":[141],"increases":[142],"block.":[146],"In":[147],"this":[148],"paper,":[149],"we":[150],"propose":[151],"<italic":[152,166,195,232,264],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[153,167,196,233,265],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">GHCoM</i>,":[154],"a":[155,236],"novel":[156],"block-wise":[157,226],"framework":[159],"exploring":[161],"effective":[163],"trade-offs.":[165],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">GHCoM</i>":[168,197,234,266],"employs":[169],"an":[170],"adaptive":[171],"strategy":[173],"guide":[175],"based":[178],"topology":[181],"DNNs":[183],"dynamically":[185],"adjust":[186],"according":[189],"target.":[193],"Furthermore,":[194],"considers":[198],"fine-grained":[200],"(i.e.,":[204,209],"processing":[205],"elements)":[206],"on-chip":[210],"bandwidth)":[211],"resources":[212],"mitigate":[217],"interlayer":[218],"contention.":[220],"To":[221],"jointly":[222],"grouping,":[225],"allocation,":[231],"leverages":[235],"two-level":[237],"genetic":[238],"algorithm":[239],"(GA)":[240],"tailored":[242],"encodings":[243],"operators":[245],"that":[246,263],"capture":[247],"interdependence":[249],"across":[250,256],"entire":[252],"design":[253],"space.":[254],"Experiments":[255],"configurations":[261],"show":[262],"consistently":[267],"outperforms":[268],"state-of-the-art":[269],"baselines,":[270],"1.08\u00d7":[272],"4.79\u00d7":[274],"speedup":[275],"latency":[278],"reducing":[280],"energy":[281],"consumption":[282],"by":[283],"1.83%":[284],"87.71%.":[286]},"counts_by_year":[],"updated_date":"2026-03-09T07:00:12.390032","created_date":"2026-02-24T00:00:00"}