{"id":"https://openalex.org/W7164150025","doi":"https://doi.org/10.1109/fccm68464.2026.00036","title":"When Systolic Arrays Meet AI Engines: Architectural Constraints on AMD Versal ACAP","display_name":"When Systolic Arrays Meet AI Engines: Architectural Constraints on AMD Versal ACAP","publication_year":2026,"publication_date":"2026-05-13","ids":{"openalex":"https://openalex.org/W7164150025","doi":"https://doi.org/10.1109/fccm68464.2026.00036"},"language":null,"primary_location":{"id":"doi:10.1109/fccm68464.2026.00036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fccm68464.2026.00036","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 34th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","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/A5070019339","display_name":"Jason Kimko","orcid":null},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jason Kimko","raw_affiliation_strings":["University of California,Los Angeles"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California,Los Angeles","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5138334489","display_name":"Jason Cong","orcid":null},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jason Cong","raw_affiliation_strings":["University of California,Los Angeles"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California,Los Angeles","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I161318765"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.85289791,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"185","last_page":"193"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.36160001158714294,"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":0.36160001158714294,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.09870000183582306,"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":0.09260000288486481,"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/software","display_name":"Software","score":0.3000999987125397},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.27489998936653137},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.27480000257492065},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.24899999797344208}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.621999979019165},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.303600013256073},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C199639397","wikidata":"https://www.wikidata.org/wiki/Q1788588","display_name":"Engineering drawing","level":1,"score":0.27790001034736633},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27489998936653137},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.25060001015663147},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.24899999797344208},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.24169999361038208},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.23810000717639923}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/fccm68464.2026.00036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fccm68464.2026.00036","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 34th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W2606722458","https://openalex.org/W3024621361","https://openalex.org/W3130920634","https://openalex.org/W4319870545","https://openalex.org/W4391455288","https://openalex.org/W4401331008","https://openalex.org/W4401568524","https://openalex.org/W4403278819","https://openalex.org/W4407953492","https://openalex.org/W4411485994","https://openalex.org/W7124680514","https://openalex.org/W7140373340"],"related_works":[],"abstract_inverted_index":{"The":[0],"AMD":[1],"Versal":[2],"ACAP":[3],"provides":[4],"a":[5,28,73,84,114,126,144,156,170,187,205],"powerful":[6],"heterogeneous":[7],"architecture":[8],"for":[9,107,135,150,189],"accelerating":[10],"kernels":[11],"like":[12],"general":[13],"matrix":[14],"multiplication":[15],"(GEMM).":[16],"Because":[17],"the":[18,24,36,61,70,79,93,163,197],"AI":[19],"Engine":[20],"(AIE)":[21],"array":[22,30],"mimics":[23],"spatial":[25],"regularity":[26],"of":[27,72,87,199],"systolic":[29,89,177,190],"(SA),":[31],"mapping":[32,90],"SA":[33,128],"dataflows":[34],"onto":[35],"fabric":[37],"is":[38,69,141,180],"highly":[39],"intuitive.":[40],"However,":[41],"maximizing":[42],"performance":[43,68,179],"requires":[44],"overcoming":[45],"specific":[46],"coordination":[47],"challenges":[48],"between":[49,63],"data":[50,100,202],"movement":[51,203],"and":[52,66,195],"compute":[53,117],"cores.":[54],"In":[55],"this":[56,122],"work,":[57],"we":[58,124],"demonstrate":[59],"that":[60,131,160],"gap":[62],"theoretical":[64],"peak":[65],"realized":[67],"result":[71],"fundamental":[74],"\"Systolic":[75],"Tax\"":[76],"imposed":[77],"by":[78,143,183],"AIE":[80],"architecture.":[81],"We":[82],"provide":[83],"systematic":[85],"study":[86],"2D":[88],"to":[91,113],"quantify":[92],"architectural":[94,151,207],"cost":[95],"incurred":[96],"when":[97],"traditional":[98],"wavefront":[99],"forwarding":[101],"consumes":[102],"execution":[103],"cycles":[104],"otherwise":[105],"available":[106],"processing.":[108],"This":[109,139],"engine-level":[110],"overhead":[111],"leads":[112],"simulated":[115],"8.2%":[116],"cycle":[118],"loss.":[119],"To":[120],"address":[121],"constraint,":[123],"implement":[125],"multicast-based":[127],"(MCSA)":[129],"design":[130,140,154],"leverages":[132],"hardware":[133],"multicasting":[134],"efficient":[136],"operand":[137],"delivery.":[138],"supported":[142],"Mixed-Integer":[145],"Non-Linear":[146],"Programming":[147],"model":[148],"used":[149],"optimization.":[152],"Our":[153],"achieves":[155],"programmable":[157],"logic":[158],"frequency":[159],"analytically":[161],"saturates":[162],"logic-engine":[164],"communication":[165],"bandwidth.":[166],"While":[167],"MCSA":[168],"delivers":[169],"1.28\u00d7":[171],"throughput":[172],"improvement":[173],"over":[174],"prior":[175],"state-of-the-art":[176],"mappings,":[178],"ultimately":[181],"limited":[182],"synchronization":[184],"overhead,":[185],"establishing":[186],"ceiling":[188],"designs":[191],"on":[192],"these":[193],"platforms":[194],"revealing":[196],"lack":[198],"parallel":[200],"AIE-initiated":[201],"as":[204],"critical":[206],"bottleneck.":[208]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2026-06-11T00:00:00"}
