{"id":"https://openalex.org/W2936567716","doi":"https://doi.org/10.1177/1094342020921340","title":"A mechanism for balancing accuracy and scope in cross-machine black-box GPU performance modeling","display_name":"A mechanism for balancing accuracy and scope in cross-machine black-box GPU performance modeling","publication_year":2020,"publication_date":"2020-06-03","ids":{"openalex":"https://openalex.org/W2936567716","doi":"https://doi.org/10.1177/1094342020921340","mag":"2936567716"},"language":"en","primary_location":{"id":"doi:10.1177/1094342020921340","is_oa":true,"landing_page_url":"https://doi.org/10.1177/1094342020921340","pdf_url":"https://journals.sagepub.com/doi/pdf/10.1177/1094342020921340","source":{"id":"https://openalex.org/S60606485","display_name":"The International Journal of High Performance Computing Applications","issn_l":"0890-2720","issn":["0890-2720","1094-3420","1741-2846"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of High Performance Computing Applications","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://journals.sagepub.com/doi/pdf/10.1177/1094342020921340","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"James D Stevens","orcid":"https://orcid.org/0000-0003-3329-3279"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"James D Stevens","raw_affiliation_strings":["Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0000-0003-3329-3279","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":null,"display_name":"Andreas Kl\u00f6ckner","orcid":null},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Andreas Kl\u00f6ckner","raw_affiliation_strings":["Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I157725225"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.01066667,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":"34","issue":"6","first_page":"589","last_page":"614"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9490000009536743,"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":0.9490000009536743,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.010499999858438969,"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":0.008700000122189522,"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/stencil","display_name":"Stencil","score":0.617900013923645},{"id":"https://openalex.org/keywords/coprocessor","display_name":"Coprocessor","score":0.6047999858856201},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5864999890327454},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5800999999046326},{"id":"https://openalex.org/keywords/massively-parallel","display_name":"Massively parallel","score":0.5616999864578247},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5329999923706055},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5092999935150146},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.47699999809265137},{"id":"https://openalex.org/keywords/parameterized-complexity","display_name":"Parameterized complexity","score":0.4002000093460083}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8604000210762024},{"id":"https://openalex.org/C76752949","wikidata":"https://www.wikidata.org/wiki/Q7607499","display_name":"Stencil","level":2,"score":0.617900013923645},{"id":"https://openalex.org/C86111242","wikidata":"https://www.wikidata.org/wiki/Q859595","display_name":"Coprocessor","level":2,"score":0.6047999858856201},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5864999890327454},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5800999999046326},{"id":"https://openalex.org/C190475519","wikidata":"https://www.wikidata.org/wiki/Q544384","display_name":"Massively parallel","level":2,"score":0.5616999864578247},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5329999923706055},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.49900001287460327},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.47699999809265137},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.4002000093460083},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.3709000051021576},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.35359999537467957},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3386000096797943},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.33809998631477356},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.3269999921321869},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.30489999055862427},{"id":"https://openalex.org/C1306188","wikidata":"https://www.wikidata.org/wiki/Q4060687","display_name":"Nested loop join","level":2,"score":0.3009999990463257},{"id":"https://openalex.org/C63000827","wikidata":"https://www.wikidata.org/wiki/Q3080428","display_name":"Software portability","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.28519999980926514},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.2849999964237213},{"id":"https://openalex.org/C172430144","wikidata":"https://www.wikidata.org/wiki/Q17111997","display_name":"Symmetric multiprocessor system","level":2,"score":0.27889999747276306},{"id":"https://openalex.org/C2777115002","wikidata":"https://www.wikidata.org/wiki/Q7168246","display_name":"Performance prediction","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C139352143","wikidata":"https://www.wikidata.org/wiki/Q82571","display_name":"Linear algebra","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.25850000977516174},{"id":"https://openalex.org/C2780799671","wikidata":"https://www.wikidata.org/wiki/Q17087362","display_name":"Transient (computer programming)","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.25679999589920044},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.25130000710487366}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1177/1094342020921340","is_oa":true,"landing_page_url":"https://doi.org/10.1177/1094342020921340","pdf_url":"https://journals.sagepub.com/doi/pdf/10.1177/1094342020921340","source":{"id":"https://openalex.org/S60606485","display_name":"The International Journal of High Performance Computing Applications","issn_l":"0890-2720","issn":["0890-2720","1094-3420","1741-2846"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of High Performance Computing Applications","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:1904.09538","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1904.09538","pdf_url":"https://arxiv.org/pdf/1904.09538","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1177/1094342020921340","is_oa":true,"landing_page_url":"https://doi.org/10.1177/1094342020921340","pdf_url":"https://journals.sagepub.com/doi/pdf/10.1177/1094342020921340","source":{"id":"https://openalex.org/S60606485","display_name":"The International Journal of High Performance Computing Applications","issn_l":"0890-2720","issn":["0890-2720","1094-3420","1741-2846"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of High Performance Computing Applications","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G557630095","display_name":null,"funder_award_id":"N00014-14-1-0117","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"}],"funders":[{"id":"https://openalex.org/F4320337345","display_name":"Office of Naval Research","ror":"https://ror.org/00rk2pe57"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W1480947737","https://openalex.org/W1558370006","https://openalex.org/W1965742121","https://openalex.org/W1967701350","https://openalex.org/W1981896890","https://openalex.org/W2021476856","https://openalex.org/W2023531503","https://openalex.org/W2038666141","https://openalex.org/W2047012348","https://openalex.org/W2072916763","https://openalex.org/W2075178432","https://openalex.org/W2091973955","https://openalex.org/W2097117297","https://openalex.org/W2099891946","https://openalex.org/W2105545110","https://openalex.org/W2109426995","https://openalex.org/W2109608899","https://openalex.org/W2113282196","https://openalex.org/W2118963820","https://openalex.org/W2120919211","https://openalex.org/W2122297544","https://openalex.org/W2129740858","https://openalex.org/W2130336316","https://openalex.org/W2130540010","https://openalex.org/W2134495056","https://openalex.org/W2147543317","https://openalex.org/W2149234156","https://openalex.org/W2152513418","https://openalex.org/W2156286566","https://openalex.org/W2167334577","https://openalex.org/W2168361773","https://openalex.org/W2170059648","https://openalex.org/W2314321304","https://openalex.org/W2343667998","https://openalex.org/W2588108287","https://openalex.org/W2607112060","https://openalex.org/W2899387410","https://openalex.org/W2982826349","https://openalex.org/W3005101099","https://openalex.org/W3006997840","https://openalex.org/W3138798301","https://openalex.org/W4240916550","https://openalex.org/W6922490038"],"related_works":[],"abstract_inverted_index":{"The":[0],"ability":[1],"to":[2,91,118,121,134,188,241],"model,":[3],"analyze,":[4],"and":[5,30,93,109,142,151,164,181,209,229,266],"predict":[6,189],"execution":[7,190],"time":[8],"of":[9,33,114,174,195,268,279],"computations":[10],"is":[11],"an":[12,77,270],"important":[13],"building":[14],"block":[15],"that":[16],"supports":[17],"numerous":[18],"efforts,":[19],"such":[20],"as":[21,158,162,165,169,238],"load":[22],"balancing,":[23],"benchmarking,":[24],"job":[25],"scheduling,":[26],"developer-guided":[27],"performance":[28,34,83,248,255],"optimization,":[29],"the":[31,67,132,153],"automation":[32],"tuning":[35],"for":[36,79,85,102,192],"high":[37],"performance,":[38],"parallel":[39,58],"applications.":[40],"In":[41],"today\u2019s":[42],"increasingly":[43,64],"heterogeneous":[44],"computing":[45],"environment,":[46],"this":[47,73,128,234],"task":[48],"must":[49],"be":[50],"accomplished":[51],"efficiently":[52],"across":[53],"multiple":[54,193,227],"architectures,":[55],"including":[56,88],"massively":[57],"coprocessors":[59],"like":[60],"GPUs,":[61],"which":[62],"are":[63,186],"prevalent":[65],"in":[66,123,246,256],"world\u2019s":[68],"fastest":[69],"supercomputers.":[70],"To":[71],"address":[72],"challenge,":[74],"we":[75,130,177,220,252,272],"present":[76,221],"approach":[78,237],"constructing":[80],"customizable,":[81],"cross-machine":[82],"models":[84,105,120],"GPU":[86,247,254],"kernels,":[87],"a":[89,100,110,124,196,239,242,257],"mechanism":[90],"automatically":[92],"symbolically":[94],"gather":[95],"performance-relevant":[96],"kernel":[97,280],"operation":[98,207],"counts,":[99,108],"tool":[101],"formulating":[103],"mathematical":[104],"using":[106],"these":[107,184],"customizable":[111],"parameterized":[112],"collection":[113,278],"benchmark":[115],"kernels":[116],"used":[117],"calibrate":[119],"GPUs":[122,225],"black-box":[125],"fashion.":[126],"With":[127],"approach,":[129,176],"empower":[131],"user":[133,145],"manage":[135],"trade-offs":[136],"between":[137],"model":[138,150,253],"accuracy,":[139,262],"evaluation":[140,263],"speed,":[141,264],"generalizability.":[143],"A":[144],"can":[146,251],"define":[147],"their":[148],"own":[149],"customize":[152],"calibration":[154],"process,":[155],"making":[156],"it":[157],"simple":[159],"or":[160,167,281],"complex":[161],"desired,":[163],"application-targeted":[166],"general":[168],"desired.":[170],"As":[171],"application":[172],"examples":[173,185],"our":[175],"demonstrate":[178],"both":[179],"linear":[180],"nonlinear":[182],"models;":[183],"designed":[187],"times":[191],"variants":[194],"particular":[197],"computation:":[198],"two":[199,210],"matrix-matrix":[200],"multiplication":[201],"variants,":[202,208],"four":[203],"discontinuous":[204],"Galerkin":[205],"differentiation":[206],"2D":[211],"five-point":[212],"finite":[213],"difference":[214],"stencil":[215],"variants.":[216],"For":[217],"each":[218],"variant,":[219],"accuracy":[222],"results":[223],"on":[224],"from":[226],"vendors":[228],"hardware":[230,282],"generations.":[231],"We":[232],"view":[233],"highly":[235],"user-customizable":[236],"response":[240],"central":[243],"question":[244],"arising":[245],"modeling:":[249],"how":[250],"cost-explanatory":[258],"fashion":[259],"while":[260],"maintaining":[261],"portability,":[265],"ease":[267],"use,":[269],"attribute":[271],"believe":[273],"precludes":[274],"approaches":[275],"requiring":[276],"manual":[277],"statistics.":[283]},"counts_by_year":[{"year":2024,"cited_by_count":1}],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2019-04-25T00:00:00"}