{"id":"https://openalex.org/W7148282898","doi":"https://doi.org/10.1109/hipcw66559.2025.00109","title":"Dissecting the NVIDIA RTX Blackwell Architecture with Microbenchmarks","display_name":"Dissecting the NVIDIA RTX Blackwell Architecture with Microbenchmarks","publication_year":2025,"publication_date":"2025-12-17","ids":{"openalex":"https://openalex.org/W7148282898","doi":"https://doi.org/10.1109/hipcw66559.2025.00109"},"language":null,"primary_location":{"id":"doi:10.1109/hipcw66559.2025.00109","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hipcw66559.2025.00109","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 32nd International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW)","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/A5047370789","display_name":"Aaron Jarmusch","orcid":"https://orcid.org/0000-0002-5532-6513"},"institutions":[{"id":"https://openalex.org/I86501945","display_name":"University of Delaware","ror":"https://ror.org/01sbq1a82","country_code":"US","type":"education","lineage":["https://openalex.org/I86501945"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aaron Jarmusch","raw_affiliation_strings":["University of Delaware,Newark,US"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Delaware,Newark,US","institution_ids":["https://openalex.org/I86501945"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132769263","display_name":"Nathan Graddon","orcid":null},"institutions":[{"id":"https://openalex.org/I86501945","display_name":"University of Delaware","ror":"https://ror.org/01sbq1a82","country_code":"US","type":"education","lineage":["https://openalex.org/I86501945"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nathan Graddon","raw_affiliation_strings":["University of Delaware,Newark,US"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Delaware,Newark,US","institution_ids":["https://openalex.org/I86501945"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5011871104","display_name":"S. Chandrasekaran","orcid":null},"institutions":[{"id":"https://openalex.org/I86501945","display_name":"University of Delaware","ror":"https://ror.org/01sbq1a82","country_code":"US","type":"education","lineage":["https://openalex.org/I86501945"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sunita Chandrasekaran","raw_affiliation_strings":["University of Delaware,Newark,US"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Delaware,Newark,US","institution_ids":["https://openalex.org/I86501945"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.75332994,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"309","last_page":"310"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.10289999842643738,"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":0.10289999842643738,"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/T12034","display_name":"Digital and Cyber Forensics","score":0.0860000029206276,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.07020000368356705,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4043999910354614},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.3000999987125397},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2565999925136566},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.24529999494552612},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.23999999463558197}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6636999845504761},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4043999910354614},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.26460000872612},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C70721500","wikidata":"https://www.wikidata.org/wiki/Q177005","display_name":"Computational biology","level":1,"score":0.24690000712871552},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.24529999494552612},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.23999999463558197},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2395000010728836},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.23880000412464142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hipcw66559.2025.00109","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hipcw66559.2025.00109","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 32nd International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW)","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":1,"referenced_works":["https://openalex.org/W2110195531"],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1],"the":[2],"first":[3],"microarchitectural":[4],"analysis":[5],"of":[6],"NVIDIA\u2019s":[7],"Blackwell":[8],"(GB203)":[9],"through":[10],"PTX":[11],"microbenchmarks,":[12],"comparing":[13],"against":[14],"Hopper":[15,36],"(GH100).":[16],"Our":[17],"measurements":[18],"reveal":[19],"Blackwell\u2019s":[20],"unified":[21],"INT32/FP32":[22],"cores":[23],"reduce":[24],"mixed-workload":[25],"latency":[26],"by":[27],"50%,":[28],"while":[29],"FP4":[30],"achieves":[31],"64%":[32],"power":[33],"reduction.":[34],"However,":[35],"maintains":[37],"4\u00d7":[38],"higher":[39],"GEMM":[40],"throughput,":[41],"exposing":[42],"significant":[43],"software":[44],"optimization":[45],"gaps.":[46],"These":[47],"findings":[48],"provide":[49],"actionable":[50],"guidelines":[51],"for":[52],"GPU":[53],"architecture":[54],"selection.":[55]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-04-03T00:00:00"}