{"id":"https://openalex.org/W4410394411","doi":"https://doi.org/10.1109/lca.2025.3570667","title":"DynaFlow: An ML Framework for Dynamic Dataflow Selection in SpGEMM Accelerators","display_name":"DynaFlow: An ML Framework for Dynamic Dataflow Selection in SpGEMM Accelerators","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4410394411","doi":"https://doi.org/10.1109/lca.2025.3570667"},"language":"en","primary_location":{"id":"doi:10.1109/lca.2025.3570667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3570667","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","raw_type":"journal-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/A5113255611","display_name":"Sanjali Yadav","orcid":null},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sanjali Yadav","raw_affiliation_strings":["University of Maryland, College Park, MD, USA","University of Maryland, College Park, USA"],"affiliations":[{"raw_affiliation_string":"University of Maryland, College Park, MD, USA","institution_ids":["https://openalex.org/I66946132"]},{"raw_affiliation_string":"University of Maryland, College Park, USA","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5059742939","display_name":"Bahar Asgari","orcid":"https://orcid.org/0000-0003-2305-9892"},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bahar Asgari","raw_affiliation_strings":["University of Maryland, College Park, MD, USA","University of Maryland, College Park, USA"],"affiliations":[{"raw_affiliation_string":"University of Maryland, College Park, MD, USA","institution_ids":["https://openalex.org/I66946132"]},{"raw_affiliation_string":"University of Maryland, College Park, USA","institution_ids":["https://openalex.org/I66946132"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5113255611"],"corresponding_institution_ids":["https://openalex.org/I66946132"],"apc_list":null,"apc_paid":null,"fwci":7.098,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.96830647,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"24","issue":"1","first_page":"189","last_page":"192"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9990000128746033,"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"}},"topics":[{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9990000128746033,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9977999925613403,"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/T11044","display_name":"Particle Detector Development and Performance","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/3106","display_name":"Nuclear and High Energy Physics"},"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/dataflow","display_name":"Dataflow","score":0.8877217173576355},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.817638635635376},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.6514531373977661},{"id":"https://openalex.org/keywords/dataflow-architecture","display_name":"Dataflow architecture","score":0.5127323865890503},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4953876733779907},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4496515095233917},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.40574902296066284},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.3685617446899414},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.32833677530288696},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.10342240333557129}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.8877217173576355},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.817638635635376},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.6514531373977661},{"id":"https://openalex.org/C176727019","wikidata":"https://www.wikidata.org/wiki/Q1172415","display_name":"Dataflow architecture","level":3,"score":0.5127323865890503},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4953876733779907},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4496515095233917},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.40574902296066284},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.3685617446899414},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.32833677530288696},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.10342240333557129}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lca.2025.3570667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3570667","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320332359","display_name":"Office of Science","ror":"https://ror.org/00mmn6b08"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":10,"referenced_works":["https://openalex.org/W2035080386","https://openalex.org/W2794952988","https://openalex.org/W3016542674","https://openalex.org/W3016832937","https://openalex.org/W3103168911","https://openalex.org/W3155922894","https://openalex.org/W3205727737","https://openalex.org/W4327930477","https://openalex.org/W4401212175","https://openalex.org/W4404955127"],"related_works":["https://openalex.org/W2564598376","https://openalex.org/W1484403103","https://openalex.org/W2584408851","https://openalex.org/W2115158825","https://openalex.org/W2101960124","https://openalex.org/W4377693460","https://openalex.org/W2033683327","https://openalex.org/W2783505431","https://openalex.org/W2521947294","https://openalex.org/W4236419692"],"abstract_inverted_index":{"Sparse":[0],"matrix-matrix":[1],"multiplication":[2],"(SpGEMM)":[3],"is":[4],"a":[5,70,98,144],"critical":[6],"operation":[7],"in":[8],"numerous":[9],"fields,":[10],"including":[11],"scientific":[12],"computing,":[13],"graph":[14],"analytics,":[15],"and":[16,26,114,131],"deep":[17,132],"learning,":[18,134],"leveraging":[19,128],"matrix":[20],"sparsity":[21,52,74,90,125],"to":[22,55,88,116,143],"reduce":[23],"both":[24],"storage":[25],"computation":[27],"costs.":[28],"However,":[29],"the":[30,59,76,105,118,123],"irregular":[31],"structure":[32],"of":[33,73,82,107],"sparse":[34],"matrices":[35],"poses":[36],"significant":[37],"challenges":[38],"for":[39,50,79],"performance":[40,56],"optimization.":[41],"Existing":[42],"hardware":[43],"accelerators":[44,80],"often":[45],"employ":[46],"fixed":[47],"dataflows":[48,108],"designed":[49],"specific":[51],"patterns,":[53],"leading":[54],"degradation":[57],"when":[58],"input":[60,124],"deviates":[61],"from":[62],"these":[63],"assumptions.":[64],"As":[65],"SpGEMM":[66],"adoption":[67],"expands":[68],"across":[69],"broad":[71],"spectrum":[72],"workloads,":[75],"demand":[77],"grows":[78],"capable":[81],"dynamically":[83],"adapting":[84],"their":[85],"dataflow":[86,120,138],"schemes":[87],"diverse":[89],"patterns.":[91],"To":[92],"address":[93],"this,":[94],"we":[95],"propose":[96],"DynaFlow,":[97],"machine":[99],"learning-based":[100],"framework":[101],"that":[102],"trains":[103],"on":[104,122],"set":[106],"supported":[109],"by":[110],"any":[111],"given":[112],"accelerator":[113],"learns":[115],"predict":[117],"optimal":[119],"based":[121],"pattern.":[126],"By":[127],"decision":[129],"trees":[130],"reinforcement":[133],"DynaFlow":[135],"surpasses":[136],"static":[137],"selection":[139],"approaches,":[140],"achieving":[141],"up":[142],"50\u00d7":[145],"speedup.":[146]},"counts_by_year":[{"year":2025,"cited_by_count":5}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}