{"id":"https://openalex.org/W7165145162","doi":"https://doi.org/10.1145/3787109.3815309","title":"\u03bc-ORCA: Optimizing Acceleration for Microsecond-Scale Deep Neural Network Inference on ACAP","display_name":"\u03bc-ORCA: Optimizing Acceleration for Microsecond-Scale Deep Neural Network Inference on ACAP","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165145162","doi":"https://doi.org/10.1145/3787109.3815309"},"language":null,"primary_location":{"id":"doi:10.1145/3787109.3815309","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815309","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3787109.3815309","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103205936","display_name":"Shixin Ji","orcid":"https://orcid.org/0009-0003-3429-4692"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shixin Ji","raw_affiliation_strings":["Brown University, Providence, RI, USA"],"raw_orcid":"https://orcid.org/0009-0003-3429-4692","affiliations":[{"raw_affiliation_string":"Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074853634","display_name":"Jinming Zhuang","orcid":"https://orcid.org/0000-0003-3659-339X"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinming Zhuang","raw_affiliation_strings":["Brown University, Providence, RI, USA"],"raw_orcid":"https://orcid.org/0000-0003-3659-339X","affiliations":[{"raw_affiliation_string":"Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088193007","display_name":"Zhuoping Yang","orcid":"https://orcid.org/0000-0002-7655-4080"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhuoping Yang","raw_affiliation_strings":["Brown University, Providence, RI, USA"],"raw_orcid":"https://orcid.org/0000-0002-7655-4080","affiliations":[{"raw_affiliation_string":"Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055054641","display_name":"X. Chen","orcid":"https://orcid.org/0000-0003-4865-3708"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xingzhen Chen","raw_affiliation_strings":["Brown University, Providence, RI, USA"],"raw_orcid":"https://orcid.org/0000-0003-4865-3708","affiliations":[{"raw_affiliation_string":"Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114193443","display_name":"Wei Zhang","orcid":"https://orcid.org/0009-0004-4590-6559"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Zhang","raw_affiliation_strings":["Brown University, Providence, USA"],"raw_orcid":"https://orcid.org/0009-0004-4590-6559","affiliations":[{"raw_affiliation_string":"Brown University, Providence, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063866156","display_name":"Peipei Zhou","orcid":"https://orcid.org/0000-0002-0493-1844"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peipei Zhou","raw_affiliation_strings":["Brown University, Providence, RI, USA"],"raw_orcid":"https://orcid.org/0000-0002-0493-1844","affiliations":[{"raw_affiliation_string":"Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I27804330"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.76967087,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"898","last_page":"905"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.382099986076355,"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":0.382099986076355,"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":0.27239999175071716,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.06539999693632126,"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/latency","display_name":"Latency (audio)","score":0.7001000046730042},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6525999903678894},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5735999941825867},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.48980000615119934},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.4697999954223633},{"id":"https://openalex.org/keywords/design-space-exploration","display_name":"Design space exploration","score":0.4007999897003174},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.39410001039505005},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.3840999901294708}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8421000242233276},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.7001000046730042},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6525999903678894},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5735999941825867},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.48980000615119934},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.4697999954223633},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.45489999651908875},{"id":"https://openalex.org/C2776221188","wikidata":"https://www.wikidata.org/wiki/Q21072556","display_name":"Design space exploration","level":2,"score":0.4007999897003174},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.39410001039505005},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3898000121116638},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.3840999901294708},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3652999997138977},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.36239999532699585},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.3540000021457672},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.3398999869823456},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31380000710487366},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.304500013589859},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2922999858856201},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.26489999890327454},{"id":"https://openalex.org/C34146451","wikidata":"https://www.wikidata.org/wiki/Q5048094","display_name":"Cascade","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3787109.3815309","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815309","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3787109.3815309","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815309","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1646487257","display_name":null,"funder_award_id":"2544032","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G1715707926","display_name":"Collaborative Research: DESC: Type II: REFRESH: Revisiting Expanding FPGA Real-estate for Environmentally Sustainability Heterogeneous-Systems","funder_award_id":"2536952","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4118315690","display_name":"MRI Consortium: Track 1: Development of SPECTRUM: An Evolutionary Rapid-Prototyping Testbed for Low Latency Multi-Domain Computing","funder_award_id":"2511445","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5405343948","display_name":"POSE: Phase I:  Enabling Open-source Ecosystem for Rapid System-on-Chip Design and Programming","funder_award_id":"2518375","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7022024765","display_name":null,"funder_award_id":"DE-SC0026344","funder_id":"https://openalex.org/F4320306084","funder_display_name":"U.S. Department of Energy"},{"id":"https://openalex.org/G7849508813","display_name":"Collaborative Research: PPoSS: LARGE: Co-designing  Hardware, Software, and Algorithms to Enable Extreme-Scale Machine Learning Systems","funder_award_id":"2348306","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/F4320306084","display_name":"U.S. Department of Energy","ror":"https://ror.org/01bj3aw27"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2951537853","https://openalex.org/W3093982621","https://openalex.org/W3103837983","https://openalex.org/W4308090436","https://openalex.org/W4386763982","https://openalex.org/W4388214748","https://openalex.org/W4389162879","https://openalex.org/W4391125526","https://openalex.org/W4391455288","https://openalex.org/W4391455313","https://openalex.org/W4391591754","https://openalex.org/W4401331008","https://openalex.org/W4403278819","https://openalex.org/W4405792756","https://openalex.org/W4407953492","https://openalex.org/W4410810249","https://openalex.org/W4411485994","https://openalex.org/W4412987090","https://openalex.org/W4413097994","https://openalex.org/W4415939076","https://openalex.org/W4416296600","https://openalex.org/W7110310012","https://openalex.org/W7117687032","https://openalex.org/W7140373340","https://openalex.org/W7164131799","https://openalex.org/W7164202102"],"related_works":[],"abstract_inverted_index":{"Heterogeneous":[0],"reconfigurable":[1],"platforms":[2],"with":[3,160,198],"tensor":[4],"cores,":[5],"such":[6,64],"as":[7,65],"AMD":[8,176],"ACAP,":[9],"are":[10,71],"increasingly":[11],"adopted":[12],"for":[13,29,93,209],"deep":[14],"neural":[15],"network":[16],"(DNN)":[17],"inference":[18,31],"due":[19],"to":[20,77,140,150],"their":[21,27],"high":[22],"throughput":[23],"and":[24,47,67,145,157,166,193,203],"flexibility.":[25],"However,":[26],"suitability":[28],"microsecond-scale":[30],"on":[32,105,169,174],"small":[33],"problem":[34],"sizes":[35],"remains":[36],"underexplored.":[37],"In":[38],"jet-tagging":[39],"applications":[40],"in":[41],"high-energy":[42],"physics,":[43],"inefficient":[44],"on-chip":[45],"communication":[46,101],"large":[48],"inter-layer":[49,100],"latency":[50,59,187,208,217],"prevent":[51],"existing":[52],"frameworks":[53],"from":[54],"meeting":[55],"the":[56,106,175,216],"1-\u03bc":[57],"s":[58,207],"budget.":[60,218],"Moreover,":[61,118],"hardware":[62],"overheads":[63],"synchronization":[66],"VLIW":[68],"processor":[69],"prologue":[70],"often":[72],"overlooked,":[73],"making":[74],"it":[75],"infeasible":[76],"optimize":[78,151],"accelerators":[79],"correctly.":[80],"To":[81],"address":[82],"these":[83],"problems,":[84],"we":[85],"propose":[86],"\u00b5-ORCA,":[87],"a":[88,119,127,210],"customized":[89],"heterogeneous":[90],"accelerator":[91],"framework":[92],"ultra-low-latency":[94],"model":[95,137],"inference.":[96],"\u00b5-ORCA":[97,131,154,173,184,222],"enables":[98],"direct":[99],"between":[102],"DNN":[103],"layers":[104],"AIE":[107],"array,":[108],"instead":[109,125],"of":[110,126,189],"using":[111],"shared":[112],"memory":[113],"tiles":[114],"or":[115],"FPGA":[116],"fabric.":[117],"512-bit/cycle":[120],"cascade":[121],"connection":[122],"is":[123],"applied":[124],"32-bit/cycle":[128],"DMA":[129],"connection.":[130],"also":[132],"provides":[133],"an":[134],"overhead-aware":[135],"performance":[136],"that":[138,183],"adapts":[139],"different":[141,199],"NN":[142],"layer":[143],"sizes,":[144],"conducts":[146],"design":[147],"space":[148],"exploration":[149],"end-to-end":[152],"latency.":[153],"supports":[155],"MLP":[156],"DeepSets":[158,213],"models":[159],"non-MM":[161],"kernels,":[162],"including":[163],"bias,":[164],"ReLU,":[165],"global":[167],"aggregation":[168],"AIE.":[170],"We":[171,219],"evaluate":[172],"ACAP":[177,201],"VEK280":[178],"platform.":[179],"Experimental":[180],"results":[181],"show":[182],"achieves":[185,204],"average":[186],"reduction":[188],">":[190,194],"1.70":[191],"\u00d7":[192,196],"1.83":[195],"compared":[197],"state-of-the-art":[200],"frameworks,":[202],"0.93":[205],"\u03bc":[206],"6-layer":[211],"real-world":[212],"model,":[214],"satisfying":[215],"open":[220],"source":[221],"at":[223],"https://github.com/arc-research-lab/u-ORCA.":[224]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-06-19T00:00:00"}