{"id":"https://openalex.org/W7140373340","doi":"https://doi.org/10.1109/fpl68686.2025.00051","title":"GAMA: High-Performance GEMM Acceleration on AMD Versal ML-Optimized AI Engines","display_name":"GAMA: High-Performance GEMM Acceleration on AMD Versal ML-Optimized AI Engines","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W7140373340","doi":"https://doi.org/10.1109/fpl68686.2025.00051"},"language":null,"primary_location":{"id":"doi:10.1109/fpl68686.2025.00051","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fpl68686.2025.00051","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 35th International Conference on Field-Programmable Logic and Applications (FPL)","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/A5001383565","display_name":"Kaustubh Manohar Mhatre","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Kaustubh Manohar Mhatre","raw_affiliation_strings":["Arizona State University,Tempe,USA"],"affiliations":[{"raw_affiliation_string":"Arizona State University,Tempe,USA","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084140879","display_name":"Endri Taka","orcid":null},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Endri Taka","raw_affiliation_strings":["The University of Texas at Austin Austin,USA"],"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin Austin,USA","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5130631891","display_name":"Aman Arora","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aman Arora","raw_affiliation_strings":["Arizona State University,Tempe,USA"],"affiliations":[{"raw_affiliation_string":"Arizona State University,Tempe,USA","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5001383565"],"corresponding_institution_ids":["https://openalex.org/I55732556"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.80621935,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"323","last_page":"331"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.12449999898672104,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.12449999898672104,"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":0.11129999905824661,"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":0.08370000123977661,"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/acceleration","display_name":"Acceleration","score":0.6632000207901001},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2989000082015991},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.23109999299049377},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.2303999960422516},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.22669999301433563}],"concepts":[{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.6632000207901001},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5652999877929688},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.36160001158714294},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2989000082015991},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2678000032901764},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.23109999299049377},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.2303999960422516},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.22669999301433563},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.22450000047683716},{"id":"https://openalex.org/C89805583","wikidata":"https://www.wikidata.org/wiki/Q192940","display_name":"Accelerometer","level":2,"score":0.22139999270439148}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/fpl68686.2025.00051","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fpl68686.2025.00051","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 35th International Conference on Field-Programmable Logic and Applications (FPL)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8128347992897034}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2055312318","https://openalex.org/W3117440043","https://openalex.org/W4319870545","https://openalex.org/W4376606035","https://openalex.org/W4381328365","https://openalex.org/W4385488663","https://openalex.org/W4401331008","https://openalex.org/W4403278819","https://openalex.org/W4404102506","https://openalex.org/W4407953492","https://openalex.org/W4411485994"],"related_works":[],"abstract_inverted_index":{"General":[0],"matrix-matrix":[1],"multiplication":[2],"(GEMM)":[3],"is":[4,29],"a":[5,51],"fundamental":[6],"operation":[7],"in":[8,86],"machine":[9],"learning":[10],"(ML)":[11],"applications.":[12,34],"We":[13,49,97],"present":[14],"the":[15,64,71,93,99,114,168],"first":[16],"comprehensive":[17],"performance":[18,76,88,100],"acceleration":[19],"of":[20,102,111,132,141],"GEMM":[21,146],"workloads":[22],"on":[23],"AMD's":[24,95],"second-generation":[25],"AIE-ML":[26,39,56,65,72,116],"architecture,":[27],"which":[28],"specifically":[30],"optimized":[31],"for":[32,134,143],"ML":[33],"Compared":[35],"to":[36,92,124,167],"AI-Engine":[37],"(AIE),":[38],"offers":[40],"increased":[41],"compute":[42],"throughput":[43,164],"and":[44,66,82,89,113,137,159,175],"larger":[45],"on-chip":[46],"memory":[47,57,80],"capacity.":[48],"propose":[50],"novel":[52],"design":[53,104],"that":[54],"maximizes":[55],"utilization,":[58],"incorporates":[59],"custom":[60],"buffer":[61],"placement":[62,69],"within":[63],"staggered":[67],"kernel":[68],"across":[70],"array,":[73],"significantly":[74],"reducing":[75],"bottlenecks":[77],"such":[78],"as":[79],"stalls":[81],"routing":[83],"congestion,":[84],"resulting":[85],"improved":[87],"efficiency":[90,165],"compared":[91,166],"default":[94],"compiler.":[96],"evaluate":[98],"benefits":[101],"our":[103],"at":[105],"three":[106],"levels:":[107],"single":[108],"AIEML,":[109],"pack":[110],"AIE-ML's":[112],"complete":[115],"array.":[117],"GAMA":[118],"achieves":[119,150],"state-of-the-art":[120,169],"performance,":[121],"delivering":[122],"up":[123],"<tex":[125,151],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[126,152],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathbf{1":[127],"6":[128],"5}$</tex>":[129],"TOPS":[130],"(85%":[131],"peak)":[133,142],"int8":[135],"precision":[136,145],"83":[138],"TBFLOPS":[139],"(86%":[140],"bfloat16":[144],"workloads.":[147],"Our":[148],"solution":[149],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$8.7":[153],"\\%,":[154,156],"9":[155],"39":[157],"\\%$</tex>":[158],"53.6":[160],"%":[161],"higher":[162],"peak":[163],"AIE":[170],"frameworks":[171],"AMA,":[172],"MAXEVA,":[173],"ARIES":[174],"CHARM,":[176],"respectively.":[177]},"counts_by_year":[],"updated_date":"2026-03-28T06:11:35.319607","created_date":"2026-03-27T00:00:00"}