{"id":"https://openalex.org/W4318603355","doi":"https://doi.org/10.1109/p3hpc56579.2022.00008","title":"Understanding Strong Scaling on GPUs Using Empirical Performance Saturation Size","display_name":"Understanding Strong Scaling on GPUs Using Empirical Performance Saturation Size","publication_year":2022,"publication_date":"2022-11-01","ids":{"openalex":"https://openalex.org/W4318603355","doi":"https://doi.org/10.1109/p3hpc56579.2022.00008"},"language":"en","primary_location":{"id":"doi:10.1109/p3hpc56579.2022.00008","is_oa":false,"landing_page_url":"https://doi.org/10.1109/p3hpc56579.2022.00008","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5068047319","display_name":"David Eberius","orcid":"https://orcid.org/0000-0001-7013-7350"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"facility","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Eberius","raw_affiliation_strings":["Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","institution_ids":["https://openalex.org/I1289243028"]},{"raw_affiliation_string":"National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA","institution_ids":["https://openalex.org/I1289243028"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063162015","display_name":"Philip C. Roth","orcid":"https://orcid.org/0000-0001-9583-1103"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"facility","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Philip Roth","raw_affiliation_strings":["Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","institution_ids":["https://openalex.org/I1289243028"]},{"raw_affiliation_string":"National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA","institution_ids":["https://openalex.org/I1289243028"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012560979","display_name":"David Rogers","orcid":"https://orcid.org/0000-0002-5187-1768"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"facility","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David M. Rogers","raw_affiliation_strings":["Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Oak Ridge National Laboratory,National Center for Computational Sciences,Oak Ridge,USA","institution_ids":["https://openalex.org/I1289243028"]},{"raw_affiliation_string":"National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, USA","institution_ids":["https://openalex.org/I1289243028"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1289243028"],"apc_list":null,"apc_paid":null,"fwci":0.9278,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.72525849,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"26","last_page":"35"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":1.0,"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":1.0,"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":0.9987999796867371,"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":0.9986000061035156,"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/computer-science","display_name":"Computer science","score":0.7966457009315491},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6622107028961182},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.6127119064331055},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5746027231216431},{"id":"https://openalex.org/keywords/upper-and-lower-bounds","display_name":"Upper and lower bounds","score":0.5723551511764526},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.5186302065849304},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.5143352746963501},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.4965701699256897},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.4120291769504547},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3385747969150543},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.15030765533447266},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11976921558380127},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.10241124033927917}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7966457009315491},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6622107028961182},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.6127119064331055},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5746027231216431},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.5723551511764526},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5186302065849304},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.5143352746963501},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.4965701699256897},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.4120291769504547},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3385747969150543},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.15030765533447266},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11976921558380127},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.10241124033927917},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/p3hpc56579.2022.00008","is_oa":false,"landing_page_url":"https://doi.org/10.1109/p3hpc56579.2022.00008","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.5799999833106995,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306084","display_name":"U.S. Department of Energy","ror":"https://ror.org/01bj3aw27"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":7,"referenced_works":["https://openalex.org/W2002555321","https://openalex.org/W2913407901","https://openalex.org/W2983865192","https://openalex.org/W2994762488","https://openalex.org/W3015498000","https://openalex.org/W4200537641","https://openalex.org/W6736397125"],"related_works":["https://openalex.org/W1571368810","https://openalex.org/W1997955449","https://openalex.org/W1784146144","https://openalex.org/W2349212383","https://openalex.org/W2379400621","https://openalex.org/W2372170743","https://openalex.org/W1558545464","https://openalex.org/W2031026393","https://openalex.org/W2788783605","https://openalex.org/W2750565137"],"abstract_inverted_index":{"The":[0,28],"roofline":[1,65,80,131],"model":[2,81],"provides":[3],"a":[4,13,18,111,138],"concise":[5],"overview":[6],"of":[7,12,20,31,57,113,141],"the":[8,55],"maximum":[9],"performance":[10,26,42,66,115,122,143],"capabilities":[11],"given":[14],"computer":[15],"system":[16],"through":[17],"combination":[19],"peak":[21],"memory":[22],"bandwidth":[23],"and":[24,33,60,71,88,126],"compute":[25],"rates.":[27],"increasing":[29],"complexity":[30],"scheduling":[32],"cache":[34],"in":[35],"recent":[36,146],"GPUs,":[37],"however,":[38],"has":[39],"introduced":[40],"complicated":[41],"variability":[43],"that":[44,82,90],"is":[45],"not":[46],"captured":[47],"by":[48,98],"arithmetic":[49],"intensity":[50],"alone.":[51],"This":[52],"work":[53],"examines":[54],"effect":[56],"problem":[58,84,100,107,117],"size":[59,85,118],"GPU":[61,114,147],"launch":[62],"configurations":[63],"on":[64,93],"for":[67],"V100,":[68],"A100,":[69],"MI100,":[70],"MI250X":[72],"graphics":[73],"processing":[74],"units.":[75],"We":[76],"introduce":[77],"an":[78],"extended":[79,130],"takes":[83],"into":[86,119],"account,":[87],"find":[89],"strong":[91],"scaling":[92],"GPUs":[94],"can":[95],"be":[96],"characterized":[97],"saturation":[99],"sizes":[101,108],"as":[102],"additional":[103],"key":[104],"metrics.":[105],"Saturation":[106],"break":[109],"up":[110],"plot":[112],"vs.":[116],"three":[120],"distinct":[121],"regimes\u2013":[123],"size-limited,":[124],"cache-bound,":[125],"DRAM-bound.":[127],"With":[128],"our":[129],"model,":[132],"we":[133],"are":[134],"able":[135],"to":[136],"provide":[137],"robust":[139],"view":[140],"these":[142],"regimes":[144],"across":[145],"architectures.":[148]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}