{"id":"https://openalex.org/W4304140716","doi":"https://doi.org/10.1109/socc56010.2022.9908101","title":"CNN Implementation and Analysis on Xilinx Versal ACAP at European XFEL","display_name":"CNN Implementation and Analysis on Xilinx Versal ACAP at European XFEL","publication_year":2022,"publication_date":"2022-09-05","ids":{"openalex":"https://openalex.org/W4304140716","doi":"https://doi.org/10.1109/socc56010.2022.9908101"},"language":"en","primary_location":{"id":"doi:10.1109/socc56010.2022.9908101","is_oa":false,"landing_page_url":"https://doi.org/10.1109/socc56010.2022.9908101","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE 35th International System-on-Chip Conference (SOCC)","raw_type":"proceedings-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/A5078311452","display_name":"Ahmad Al-Zoubi","orcid":null},"institutions":[{"id":"https://openalex.org/I159176309","display_name":"Universit\u00e4t Hamburg","ror":"https://ror.org/00g30e956","country_code":"DE","type":"education","lineage":["https://openalex.org/I159176309"]},{"id":"https://openalex.org/I884043246","display_name":"Hamburg University of Technology","ror":"https://ror.org/04bs1pb34","country_code":"DE","type":"education","lineage":["https://openalex.org/I884043246"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Ahmad Al-Zoubi","raw_affiliation_strings":["Hamburg University of Technology","Center for Data and Computing in Natural Sciences"],"affiliations":[{"raw_affiliation_string":"Hamburg University of Technology","institution_ids":["https://openalex.org/I159176309","https://openalex.org/I884043246"]},{"raw_affiliation_string":"Center for Data and Computing in Natural Sciences","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036523029","display_name":"Gianluca Martino","orcid":"https://orcid.org/0000-0002-7838-3844"},"institutions":[{"id":"https://openalex.org/I2801876189","display_name":"Deutsches Elektronen-Synchrotron DESY","ror":"https://ror.org/01js2sh04","country_code":"DE","type":"facility","lineage":["https://openalex.org/I1305996414","https://openalex.org/I2801876189"]},{"id":"https://openalex.org/I159176309","display_name":"Universit\u00e4t Hamburg","ror":"https://ror.org/00g30e956","country_code":"DE","type":"education","lineage":["https://openalex.org/I159176309"]},{"id":"https://openalex.org/I884043246","display_name":"Hamburg University of Technology","ror":"https://ror.org/04bs1pb34","country_code":"DE","type":"education","lineage":["https://openalex.org/I884043246"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Gianluca Martino","raw_affiliation_strings":["Hamburg University of Technology","Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany"],"affiliations":[{"raw_affiliation_string":"Hamburg University of Technology","institution_ids":["https://openalex.org/I159176309","https://openalex.org/I884043246"]},{"raw_affiliation_string":"Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany","institution_ids":["https://openalex.org/I2801876189"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003613075","display_name":"Fin Hendrik Bahnsen","orcid":"https://orcid.org/0000-0002-5204-4713"},"institutions":[{"id":"https://openalex.org/I159176309","display_name":"Universit\u00e4t Hamburg","ror":"https://ror.org/00g30e956","country_code":"DE","type":"education","lineage":["https://openalex.org/I159176309"]},{"id":"https://openalex.org/I884043246","display_name":"Hamburg University of Technology","ror":"https://ror.org/04bs1pb34","country_code":"DE","type":"education","lineage":["https://openalex.org/I884043246"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Fin H. Bahnsen","raw_affiliation_strings":["Hamburg University of Technology"],"affiliations":[{"raw_affiliation_string":"Hamburg University of Technology","institution_ids":["https://openalex.org/I159176309","https://openalex.org/I884043246"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077498457","display_name":"Jun Zhu","orcid":"https://orcid.org/0000-0002-2544-3399"},"institutions":[{"id":"https://openalex.org/I2801876189","display_name":"Deutsches Elektronen-Synchrotron DESY","ror":"https://ror.org/01js2sh04","country_code":"DE","type":"facility","lineage":["https://openalex.org/I1305996414","https://openalex.org/I2801876189"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jun Zhu","raw_affiliation_strings":["Deutsches Elektronen-Synchrotron DESY,Hamburg,Germany,22607"],"affiliations":[{"raw_affiliation_string":"Deutsches Elektronen-Synchrotron DESY,Hamburg,Germany,22607","institution_ids":["https://openalex.org/I2801876189"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074408185","display_name":"H. Schlarb","orcid":"https://orcid.org/0000-0003-4115-5183"},"institutions":[{"id":"https://openalex.org/I2801876189","display_name":"Deutsches Elektronen-Synchrotron DESY","ror":"https://ror.org/01js2sh04","country_code":"DE","type":"facility","lineage":["https://openalex.org/I1305996414","https://openalex.org/I2801876189"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Holger Schlarb","raw_affiliation_strings":["Center for Data and Computing in Natural Sciences","Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany"],"affiliations":[{"raw_affiliation_string":"Center for Data and Computing in Natural Sciences","institution_ids":[]},{"raw_affiliation_string":"Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany","institution_ids":["https://openalex.org/I2801876189"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008766097","display_name":"G\u00f6rschwin Fey","orcid":"https://orcid.org/0000-0001-6433-6265"},"institutions":[{"id":"https://openalex.org/I884043246","display_name":"Hamburg University of Technology","ror":"https://ror.org/04bs1pb34","country_code":"DE","type":"education","lineage":["https://openalex.org/I884043246"]},{"id":"https://openalex.org/I159176309","display_name":"Universit\u00e4t Hamburg","ror":"https://ror.org/00g30e956","country_code":"DE","type":"education","lineage":["https://openalex.org/I159176309"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Goerschwin Fey","raw_affiliation_strings":["Hamburg University of Technology","Center for Data and Computing in Natural Sciences"],"affiliations":[{"raw_affiliation_string":"Hamburg University of Technology","institution_ids":["https://openalex.org/I159176309","https://openalex.org/I884043246"]},{"raw_affiliation_string":"Center for Data and Computing in Natural Sciences","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5078311452"],"corresponding_institution_ids":["https://openalex.org/I159176309","https://openalex.org/I884043246"],"apc_list":null,"apc_paid":null,"fwci":0.8053,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.72964461,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994000196456909,"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":0.9994000196456909,"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":0.9968000054359436,"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/T11949","display_name":"Nuclear Physics and Applications","score":0.9909999966621399,"subfield":{"id":"https://openalex.org/subfields/3108","display_name":"Radiation"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8037593364715576},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.798930823802948},{"id":"https://openalex.org/keywords/double-precision-floating-point-format","display_name":"Double-precision floating-point format","score":0.5263539552688599},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.5172901153564453},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.47258460521698},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4699293076992035},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4535934627056122},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.39367660880088806},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.39289426803588867},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3781816363334656},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.37797850370407104},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2602348327636719},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.240151047706604}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8037593364715576},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.798930823802948},{"id":"https://openalex.org/C35912277","wikidata":"https://www.wikidata.org/wiki/Q1243369","display_name":"Double-precision floating-point format","level":3,"score":0.5263539552688599},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.5172901153564453},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.47258460521698},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4699293076992035},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4535934627056122},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.39367660880088806},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.39289426803588867},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3781816363334656},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.37797850370407104},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2602348327636719},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.240151047706604},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/socc56010.2022.9908101","is_oa":false,"landing_page_url":"https://doi.org/10.1109/socc56010.2022.9908101","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE 35th International System-on-Chip Conference (SOCC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.44999998807907104,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2517617279","https://openalex.org/W2616014673","https://openalex.org/W2773756801","https://openalex.org/W2981712449","https://openalex.org/W2990670055","https://openalex.org/W2997593220","https://openalex.org/W3008905965","https://openalex.org/W3088552745","https://openalex.org/W3095020918","https://openalex.org/W3096235482","https://openalex.org/W3100768276","https://openalex.org/W3189184252","https://openalex.org/W3201898136","https://openalex.org/W3204657531","https://openalex.org/W4221154930"],"related_works":["https://openalex.org/W3215589575","https://openalex.org/W3150959508","https://openalex.org/W2336476964","https://openalex.org/W4297795876","https://openalex.org/W1571090276","https://openalex.org/W2773283032","https://openalex.org/W2012407419","https://openalex.org/W2239119680","https://openalex.org/W1973800584","https://openalex.org/W3150370983"],"abstract_inverted_index":{"Developers":[0],"have":[1],"proposed":[2],"various":[3],"hardware":[4],"accelerators":[5],"to":[6,33,79,101,186,204],"improve":[7],"the":[8,23,34,44,61,64,69,89,94,114,118,131,135,143,154,164,172,193,196,205],"CNN":[9,75,90,197],"inference":[10],"performance":[11,72,100,152],"on":[12],"embedded":[13],"platforms.":[14],"Recently,":[15],"Xilinx":[16,62],"announced":[17],"its":[18,99],"first":[19],"7-nm":[20],"FPGA":[21,107],"accelerator,":[22],"Versal":[24,70,155],"ACAP,":[25],"delivering":[26],"a":[27,102,127,150,209],"high-performance,":[28],"heterogeneous":[29],"computing":[30],"platform":[31],"adaptable":[32],"application":[35],"requirements.":[36],"However,":[37],"as":[38],"early":[39],"studies":[40],"were":[41,168],"concerned":[42],"with":[43,73,208],"most":[45],"common":[46],"deep":[47],"learning":[48],"architectures":[49,76,91],"for":[50,121,141],"CNN,":[51],"e.g.":[52],"VGG,":[53],"Resnet,":[54],"Inception,":[55],"etc.,":[56],"under":[57],"full":[58],"support":[59],"of":[60,68,88,116,130,153,159,211],"Vitis-AI,":[63],"implementation":[65],"and":[66,97,105,125,161,179,183],"analysis":[67,129],"ACAP":[71,156,173],"customized":[74],"is":[77],"yet":[78],"be":[80],"explored.":[81],"In":[82,109,188],"this":[83,111],"study,":[84],"we":[85],"implement":[86],"one":[87],"considered":[92],"at":[93],"European":[95],"XFEL":[96],"compare":[98],"state-of-the-art":[103],"GPU":[104],"other":[106],"generation.":[108],"addition,":[110,189],"study":[112],"evaluates":[113],"validity":[115],"using":[117,192],"quantization":[119],"methods":[120],"critical":[122],"regression":[123],"applications":[124],"presents":[126],"complete":[128],"results":[132,148],"built":[133],"upon":[134],"device":[136],"time":[137],"traces,":[138],"providing":[139],"recommendations":[140],"configuring":[142],"runtime":[144],"parameters.":[145],"The":[146],"experimental":[147],"confirm":[149],"superior":[151],"in":[157,213],"terms":[158],"latency":[160,184],"throughput.":[162],"When":[163],"neural":[165],"network":[166],"layers":[167],"all":[169],"supported":[170],"by":[171],"processing":[174],"unit,":[175],"it":[176],"achieved":[177],"17x":[178],"18x":[180],"better":[181],"throughput":[182],"compared":[185,203],"GPU.":[187],"when":[190],"quantized":[191],"fine-tuning":[194],"method,":[195],"model":[198],"shows":[199],"an":[200],"improved":[201],"accuracy":[202],"floating-point":[206],"model,":[207],"reduction":[210],"6%":[212],"loss.":[214]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}