{"id":"https://openalex.org/W4416003920","doi":"https://doi.org/10.1145/3731599.3767487","title":"The MALL is Open: Exploring Shared Caches and Latency in AMD CDNA\u2122 3 GPUs","display_name":"The MALL is Open: Exploring Shared Caches and Latency in AMD CDNA\u2122 3 GPUs","publication_year":2025,"publication_date":"2025-11-07","ids":{"openalex":"https://openalex.org/W4416003920","doi":"https://doi.org/10.1145/3731599.3767487"},"language":null,"primary_location":{"id":"doi:10.1145/3731599.3767487","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767487","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3731599.3767487","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120297507","display_name":"Andrew Tee","orcid":"https://orcid.org/0009-0001-1077-5649"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]},{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Andrew Tee","raw_affiliation_strings":["University of California, Riverside, Riverside, California, USA and Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA"],"raw_orcid":"https://orcid.org/0009-0001-1077-5649","affiliations":[{"raw_affiliation_string":"University of California, Riverside, Riverside, California, USA and Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA","institution_ids":["https://openalex.org/I4210137977","https://openalex.org/I103635307"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019774070","display_name":"Nicholas Curtis","orcid":"https://orcid.org/0000-0002-0303-4711"},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nicholas Curtis","raw_affiliation_strings":["Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA"],"raw_orcid":"https://orcid.org/0000-0002-0303-4711","affiliations":[{"raw_affiliation_string":"Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA","institution_ids":["https://openalex.org/I4210137977"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076196719","display_name":"Noah Wolfe","orcid":"https://orcid.org/0000-0002-7935-421X"},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Noah Wolfe","raw_affiliation_strings":["Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA"],"raw_orcid":"https://orcid.org/0000-0002-7935-421X","affiliations":[{"raw_affiliation_string":"Advanced Micro Devices, Inc. (AMD), Austin, Texas, USA","institution_ids":["https://openalex.org/I4210137977"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5000712719","display_name":"Daniel Wong","orcid":"https://orcid.org/0000-0002-5376-7868"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Daniel Wong","raw_affiliation_strings":["University of California, Riverside, Riverside, California, USA"],"raw_orcid":"https://orcid.org/0000-0002-5376-7868","affiliations":[{"raw_affiliation_string":"University of California, Riverside, Riverside, California, USA","institution_ids":["https://openalex.org/I103635307"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5120297507"],"corresponding_institution_ids":["https://openalex.org/I103635307","https://openalex.org/I4210137977"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.3488946,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1110","last_page":"1116"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9401000142097473,"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":0.9401000142097473,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.01759999990463257,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.014800000004470348,"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/latency","display_name":"Latency (audio)","score":0.8055999875068665},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.6718000173568726},{"id":"https://openalex.org/keywords/cas-latency","display_name":"CAS latency","score":0.6033999919891357},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.4625000059604645},{"id":"https://openalex.org/keywords/cpu-cache","display_name":"CPU cache","score":0.448199987411499},{"id":"https://openalex.org/keywords/shared-memory","display_name":"Shared memory","score":0.35499998927116394}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8242999911308289},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.8055999875068665},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.6718000173568726},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6105999946594238},{"id":"https://openalex.org/C189930140","wikidata":"https://www.wikidata.org/wiki/Q1112878","display_name":"CAS latency","level":4,"score":0.6033999919891357},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.4625000059604645},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.448199987411499},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.35499998927116394},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.35409998893737793},{"id":"https://openalex.org/C116007543","wikidata":"https://www.wikidata.org/wiki/Q1071403","display_name":"Translation lookaside buffer","level":4,"score":0.32910001277923584},{"id":"https://openalex.org/C51290061","wikidata":"https://www.wikidata.org/wiki/Q1936765","display_name":"Uniform memory access","level":4,"score":0.2953999936580658},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.27570000290870667},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.25519999861717224},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.2551000118255615},{"id":"https://openalex.org/C133371097","wikidata":"https://www.wikidata.org/wiki/Q868014","display_name":"Non-uniform memory access","level":5,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3731599.3767487","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767487","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3731599.3767487","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767487","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2011176260","display_name":null,"funder_award_id":"2047521, 1955650","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":2,"referenced_works":["https://openalex.org/W2110195531","https://openalex.org/W4308090436"],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"presents":[2],"an":[3],"analysis":[4],"of":[5],"memory":[6,141],"hierarchy":[7],"latency":[8,48,66,79,102],"across":[9,82],"AMD":[10,29,59],"Instinct\u2122":[11],"MI300A,":[12],"MI300X,":[13,55],"and":[14,41,50,54,109,118,144],"MI250X":[15,69],"GPUs":[16],"using":[17],"a":[18,71],"fine-grained":[19],"pointer-chasing":[20],"microbenchmark.":[21],"We":[22,113],"characterize":[23],"the":[24,35,58],"scalar":[25],"L1":[26],"(sL1),":[27],"L2,":[28],"Infinity":[30],"Cache\u2122":[31],"referred":[32],"to":[33,105],"as":[34],"MALL":[36,108,116,131],"(Memory":[37],"Attached":[38],"Last":[39],"Level),":[40],"HBM":[42],"(High":[43],"Bandwidth":[44],"Memory),":[45],"revealing":[46],"distinct":[47],"levels":[49],"architectural":[51],"trade-offs.":[52],"MI300A":[53],"based":[56],"on":[57,147],"CDNA\u2122":[60],"3":[61],"architecture,":[62],"exhibit":[63],"nearly":[64],"identical":[65],"profiles,":[67],"while":[68],"lacks":[70],"MALL,":[72],"resulting":[73],"in":[74,107,111],"different":[75],"performance":[76,132,146],"characteristics.":[77],"Memory":[78],"remains":[80],"consistent":[81],"compute":[83],"partitioning":[84,98],"modes,":[85],"but":[86],"NUMA":[87],"Partitioning":[88],"per":[89],"Socket":[90],"(NPS)":[91],"significantly":[92],"impacts":[93],"performance.":[94],"In":[95],"NPS4":[96],"mode,":[97],"improves":[99],"locality,":[100],"reducing":[101],"by":[103],"up":[104],"1.42\u00d7":[106],"1.31\u00d7":[110],"HBM.":[112],"further":[114],"analyze":[115],"contention":[117],"Translation":[119],"Lookaside":[120],"Buffer":[121],"(TLB)":[122],"behavior":[123],"under":[124],"varying":[125],"parallelism":[126],"levels,":[127],"identifying":[128],"conditions":[129],"where":[130],"degrades.":[133],"These":[134],"findings":[135],"provide":[136],"actionable":[137],"insights":[138],"for":[139],"optimizing":[140],"access":[142],"patterns":[143],"improving":[145],"AMD\u2019s":[148],"latest":[149],"GPU":[150],"architectures.":[151]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-11-07T00:00:00"}