{"id":"https://openalex.org/W7126200204","doi":"https://doi.org/10.1145/3767110.3767130","title":"Inter-APU Communication on AMD MI300A Systems via Infinity Fabric: A Deep Dive","display_name":"Inter-APU Communication on AMD MI300A Systems via Infinity Fabric: A Deep Dive","publication_year":2025,"publication_date":"2025-10-06","ids":{"openalex":"https://openalex.org/W7126200204","doi":"https://doi.org/10.1145/3767110.3767130"},"language":null,"primary_location":{"id":"doi:10.1145/3767110.3767130","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767110.3767130","pdf_url":null,"source":{"id":"https://openalex.org/S4306524191","display_name":"Proceedings of the International Symposium on Memory Systems","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"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 International Symposium on Memory Systems","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1145/3767110.3767130","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5092658397","display_name":"Gabin Schieffer","orcid":"https://orcid.org/0009-0003-6504-7109"},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Gabin Schieffer","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0009-0003-6504-7109","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037374138","display_name":"Jacob Wahlgren","orcid":"https://orcid.org/0000-0003-1669-7714"},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Jacob Wahlgren","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0000-0003-1669-7714","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124448726","display_name":"Ruimin Shi","orcid":null},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Ruimin Shi","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0009-0003-4387-367X","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005644355","display_name":"Edgar A. Le\u00f3n","orcid":"https://orcid.org/0000-0001-5805-9046"},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Edgar A Leon","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, CA, USA"],"raw_orcid":"https://orcid.org/0000-0001-5805-9046","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, CA, USA","institution_ids":["https://openalex.org/I1282311441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124319614","display_name":"Roger Pearce","orcid":null},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Roger Pearce","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, CA, USA"],"raw_orcid":"https://orcid.org/0009-0000-6449-6568","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, CA, USA","institution_ids":["https://openalex.org/I1282311441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041592282","display_name":"Gokhale Maya","orcid":null},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Maya Gokhale","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, CA, USA"],"raw_orcid":"https://orcid.org/0000-0003-4229-5735","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, CA, USA","institution_ids":["https://openalex.org/I1282311441"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5122360044","display_name":"Ivy Peng","orcid":"https://orcid.org/0000-0003-4158-3583"},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Ivy Peng","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0000-0003-4158-3583","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"238","last_page":"250"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9146000146865845,"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":0.9146000146865845,"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":0.022199999541044235,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.011900000274181366,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/infinity","display_name":"Infinity","score":0.7696999907493591},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5929999947547913},{"id":"https://openalex.org/keywords/data-processing","display_name":"Data processing","score":0.3411000072956085},{"id":"https://openalex.org/keywords/data-processing-system","display_name":"Data processing system","score":0.33399999141693115},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.30730000138282776},{"id":"https://openalex.org/keywords/h-infinity-methods-in-control-theory","display_name":"H-infinity methods in control theory","score":0.30219998955726624}],"concepts":[{"id":"https://openalex.org/C7321624","wikidata":"https://www.wikidata.org/wiki/Q205","display_name":"Infinity","level":2,"score":0.7696999907493591},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6377999782562256},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5929999947547913},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.40639999508857727},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3756999969482422},{"id":"https://openalex.org/C138827492","wikidata":"https://www.wikidata.org/wiki/Q6661985","display_name":"Data processing","level":2,"score":0.3411000072956085},{"id":"https://openalex.org/C112118009","wikidata":"https://www.wikidata.org/wiki/Q4925872","display_name":"Data processing system","level":2,"score":0.33399999141693115},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.30880001187324524},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.30730000138282776},{"id":"https://openalex.org/C6727046","wikidata":"https://www.wikidata.org/wiki/Q1293742","display_name":"H-infinity methods in control theory","level":3,"score":0.30219998955726624},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.29589998722076416},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.2906000018119812},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2741999924182892},{"id":"https://openalex.org/C557945733","wikidata":"https://www.wikidata.org/wiki/Q389772","display_name":"Data transmission","level":2,"score":0.26899999380111694},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.26170000433921814},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.26100000739097595},{"id":"https://openalex.org/C31352089","wikidata":"https://www.wikidata.org/wiki/Q3750474","display_name":"Systems design","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.257999986410141},{"id":"https://openalex.org/C122637931","wikidata":"https://www.wikidata.org/wiki/Q118084","display_name":"Unit (ring theory)","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3767110.3767130","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767110.3767130","pdf_url":null,"source":{"id":"https://openalex.org/S4306524191","display_name":"Proceedings of the International Symposium on Memory Systems","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"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 International Symposium on Memory Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3767110.3767130","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767110.3767130","pdf_url":null,"source":{"id":"https://openalex.org/S4306524191","display_name":"Proceedings of the International Symposium on Memory Systems","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"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 International Symposium on Memory Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","score":0.48380154371261597,"id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W1577137152","https://openalex.org/W2593535610","https://openalex.org/W2777078856","https://openalex.org/W2903659818","https://openalex.org/W2903901007","https://openalex.org/W2914698571","https://openalex.org/W2964324519","https://openalex.org/W3028661980","https://openalex.org/W3159109662","https://openalex.org/W4362566335","https://openalex.org/W4380874680","https://openalex.org/W4388581267","https://openalex.org/W4388662129","https://openalex.org/W4391451950","https://openalex.org/W4396814805","https://openalex.org/W4399304518","https://openalex.org/W4400681470","https://openalex.org/W4401211711","https://openalex.org/W4401408779","https://openalex.org/W4402702025","https://openalex.org/W4406157308","https://openalex.org/W4406157579","https://openalex.org/W4416429735"],"related_works":[],"abstract_inverted_index":{"The":[0],"ever-increasing":[1],"compute":[2,60],"performance":[3],"of":[4,94],"GPU":[5],"accelerators":[6],"drives":[7],"up":[8],"the":[9,79,92,100],"need":[10],"for":[11,110],"efficient":[12],"data":[13,28,83,127],"movements":[14],"within":[15,43,57],"HPC":[16,134],"applications":[17],"to":[18,25,73],"sustain":[19],"performance.":[20],"Proposed":[21],"as":[22,51],"a":[23,44,58,143],"solution":[24],"alleviate":[26],"CPU-GPU":[27],"movement,":[29,84],"AMD":[30,116],"MI300A":[31,117,145],"Accelerated":[32],"Processing":[33],"Unit":[34],"(APU)":[35],"combines":[36],"CPU,":[37],"GPU,":[38,80],"and":[39,85,99,126,137,139],"high-bandwidth":[40],"memory":[41,76],"(HBM)":[42],"single":[45,59],"physical":[46],"package.":[47],"Leadership":[48],"supercomputers,":[49],"such":[50],"El":[52],"Capitan,":[53],"group":[54],"four":[55,144],"APUs":[56],"node,":[61],"using":[62],"Infinity":[63,120],"Fabric":[64],"Interconnect.":[65],"In":[66],"this":[67],"work,":[68],"we":[69,130],"design":[70,108],"specific":[71],"benchmarks":[72],"evaluate":[74,140],"direct":[75],"access":[77],"from":[78],"explicit":[81],"inter-APU":[82,112],"collective":[86],"multi-APU":[87,115],"communication.":[88],"We":[89],"also":[90],"compare":[91],"efficiency":[93],"HIP":[95],"APIs,":[96],"MPI":[97],"routines,":[98],"GPU-specialized":[101],"RCCL":[102],"library.":[103],"Our":[104],"results":[105],"highlight":[106],"key":[107],"choices":[109],"optimizing":[111],"communication":[113],"on":[114,142],"systems":[118],"with":[119],"Fabric,":[121],"including":[122],"programming":[123],"interfaces,":[124],"allocators,":[125],"movement.":[128],"Finally,":[129],"optimize":[131],"two":[132],"real":[133],"applications,":[135],"Quicksilver":[136],"CloverLeaf,":[138],"them":[141],"APU":[146],"system.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-02-01T00:00:00"}