{"id":"https://openalex.org/W7132881173","doi":null,"title":"High Performance CNN Inference on FPGAs with High-Bandwidth Memory","display_name":"High Performance CNN Inference on FPGAs with High-Bandwidth Memory","publication_year":2024,"publication_date":"2024-11-01","ids":{"openalex":"https://openalex.org/W7132881173"},"language":null,"primary_location":{"id":"pmh:oai:utoronto.scholaris.ca:1807/140465","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/1807/140465","pdf_url":"https://utoronto.scholaris.ca/bitstreams/39789375-c2c7-41a4-9850-b16d0dab82de/download","source":{"id":"https://openalex.org/S4306400261","display_name":"TSpace (University of Toronto)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I185261750","host_organization_name":"University of Toronto","host_organization_lineage":["https://openalex.org/I185261750"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"},"volume":null,"type":"dissertation","indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://utoronto.scholaris.ca/bitstreams/39789375-c2c7-41a4-9850-b16d0dab82de/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5019628341","display_name":"Mario Doumet","orcid":"https://orcid.org/0000-0001-7211-1234","observed_orcids":["https://orcid.org/0000-0001-7211-1234"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Doumet, Mario","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5019628341"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.5976999998092651,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.5976999998092651,"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":0.37130001187324524,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.00430000014603138,"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/convolutional-neural-network-inference","display_name":"convolutional neural network inference","score":0.8949999809265137},{"id":"https://openalex.org/keywords/fpga-acceleration","display_name":"FPGA acceleration","score":0.75},{"id":"https://openalex.org/keywords/high-bandwidth-memory","display_name":"high-bandwidth memory","score":0.9750000238418579},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"memory bandwidth","score":0.6029999852180481},{"id":"https://openalex.org/keywords/hardware-compiler","display_name":"hardware compiler","score":0.5820000171661377},{"id":"https://openalex.org/keywords/resnet","display_name":"ResNet","score":0.6549999713897705},{"id":"https://openalex.org/keywords/vgg","display_name":"VGG","score":0.9869999885559082}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.9057000279426575},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.86080002784729},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6449999809265137},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6269999742507935},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5730000138282776},{"id":"https://openalex.org/C176727019","wikidata":"https://www.wikidata.org/wiki/Q1172415","display_name":"Dataflow architecture","level":3,"score":0.5302000045776367},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5203999876976013},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4821999967098236},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.4652000069618225},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.44440001249313354},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.42890000343322754},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.42289999127388},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.4203000068664551},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4198000133037567},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.38530001044273376},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.35269999504089355},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.34850001335144043},{"id":"https://openalex.org/C2777062904","wikidata":"https://www.wikidata.org/wiki/Q545406","display_name":"Toolchain","level":3,"score":0.3474999964237213},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.3458000123500824},{"id":"https://openalex.org/C67212190","wikidata":"https://www.wikidata.org/wiki/Q104851","display_name":"Firmware","level":2,"score":0.31139999628067017},{"id":"https://openalex.org/C58013763","wikidata":"https://www.wikidata.org/wiki/Q5754574","display_name":"High-level synthesis","level":3,"score":0.29980000853538513},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.2989000082015991},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29490000009536743},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.2827000021934509},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2815000116825104},{"id":"https://openalex.org/C77390884","wikidata":"https://www.wikidata.org/wiki/Q217302","display_name":"Application-specific integrated circuit","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C86111242","wikidata":"https://www.wikidata.org/wiki/Q859595","display_name":"Coprocessor","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:utoronto.scholaris.ca:1807/140465","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/1807/140465","pdf_url":"https://utoronto.scholaris.ca/bitstreams/39789375-c2c7-41a4-9850-b16d0dab82de/download","source":{"id":"https://openalex.org/S4306400261","display_name":"TSpace (University of Toronto)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I185261750","host_organization_name":"University of Toronto","host_organization_lineage":["https://openalex.org/I185261750"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"}],"best_oa_location":{"id":"pmh:oai:utoronto.scholaris.ca:1807/140465","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/1807/140465","pdf_url":"https://utoronto.scholaris.ca/bitstreams/39789375-c2c7-41a4-9850-b16d0dab82de/download","source":{"id":"https://openalex.org/S4306400261","display_name":"TSpace (University of Toronto)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I185261750","host_organization_name":"University of Toronto","host_organization_lineage":["https://openalex.org/I185261750"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/7","score":0.5837836265563965,"display_name":"Affordable and clean energy"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7132881173.pdf","grobid_xml":"https://content.openalex.org/works/W7132881173.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Field":[0],"Programmable":[1],"Gate":[2],"Arrays":[3],"(FPGAs)":[4],"can":[5,44],"achieve":[6,45],"low":[7],"latency":[8],"and":[9,96,113,124,142,147],"high":[10,46,53,100],"throughput":[11],"inference":[12],"on":[13,65,144],"Convolutional":[14],"Neural":[15],"Networks":[16],"(CNNs)":[17],"by":[18,61],"implementing":[19,32],"dataflow":[20,87,103],"accelerators":[21,43],"that":[22],"pipeline":[23],"layer-specific":[24],"hardware":[25],"to":[26,90,115,120,122,131],"implement":[27],"an":[28],"entire":[29],"network.":[30],"By":[31],"a":[33,52,85],"different":[34],"processing":[35],"element":[36],"for":[37,71],"each":[38],"CNN":[39],"layer,":[40],"these":[41],"layer-pipelined":[42,102],"compute":[47],"density":[48],"but":[49,67],"also":[50],"require":[51],"memory":[54],"bandwidth.":[55],"Traditionally":[56],"this":[57,68,81],"has":[58],"been":[59],"satisfied":[60],"storing":[62],"all":[63],"weights":[64],"chip,":[66],"is":[69],"infeasible":[70],"the":[72,110,117,127,132],"large":[73,106],"CNNs":[74],"most":[75],"in":[76],"need":[77],"of":[78,105,139],"acceleration.":[79],"In":[80],"work":[82,135],"we":[83,136],"augment":[84,109],"state-of-the-art":[86],"accelerator":[88],"(HPIPE)":[89],"leverage":[91],"both":[92],"High-Bandwidth":[93],"Memory":[94],"(HBM)":[95],"on-chip":[97],"storage,":[98],"enabling":[99],"performance":[101],"acceleration":[104],"CNNs.":[107],"We":[108],"HPIPE":[111],"architecture":[112],"compiler":[114],"choose":[116],"best":[118,133],"data":[119],"move":[121],"HBM,":[123],"automatically":[125],"generate":[126],"required":[128],"hardware.":[129],"Compared":[130],"prior":[134],"obtain":[137],"speed-ups":[138],"19.4x,":[140],"5.1x":[141],"10.5x":[143],"ResNet-18,":[145],"ResNet-50":[146],"VGG-16":[148],"respectively.":[149]},"counts_by_year":[],"updated_date":"2026-10-14T16:39:01.820549","created_date":"2026-03-01T00:00:00"}