{"id":"https://openalex.org/W2991644606","doi":"https://doi.org/10.1109/hpec.2019.8916392","title":"Increasing Accuracy of Iterative Refinement in Limited Floating-Point Arithmetic on Half-Precision Accelerators","display_name":"Increasing Accuracy of Iterative Refinement in Limited Floating-Point Arithmetic on Half-Precision Accelerators","publication_year":2019,"publication_date":"2019-09-01","ids":{"openalex":"https://openalex.org/W2991644606","doi":"https://doi.org/10.1109/hpec.2019.8916392","mag":"2991644606"},"language":"en","primary_location":{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec.2019.8916392","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false},"type":"article","type_crossref":"proceedings-article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.osti.gov/biblio/1641288","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073990539","display_name":"Piotr \u0141uszczek","orcid":"https://orcid.org/0000-0002-0089-6965"},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"funder","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Piotr Luszczek","raw_affiliation_strings":["University of Tennessee"],"affiliations":[{"raw_affiliation_string":"University of Tennessee","institution_ids":["https://openalex.org/I75027704"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052397462","display_name":"Ichitaro Yamazaki","orcid":"https://orcid.org/0000-0002-6196-2508"},"institutions":[{"id":"https://openalex.org/I4210104735","display_name":"Sandia National Laboratories","ror":"https://ror.org/01apwpt12","country_code":"US","type":"funder","lineage":["https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210104735"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ichitaro Yamazaki","raw_affiliation_strings":["Sandia National Laboratories"],"affiliations":[{"raw_affiliation_string":"Sandia National Laboratories","institution_ids":["https://openalex.org/I4210104735"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075517045","display_name":"Jack Dongarra","orcid":"https://orcid.org/0000-0003-3247-1782"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"funder","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]},{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"funder","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jack Dongarra","raw_affiliation_strings":["University of Tennessee, Oak Ridge National Laboratory, University of Manchester"],"affiliations":[{"raw_affiliation_string":"University of Tennessee, Oak Ridge National Laboratory, University of Manchester","institution_ids":["https://openalex.org/I1289243028","https://openalex.org/I75027704"]}]}],"institution_assertions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.77,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.593814,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":76,"max":79},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9998,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.9998,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11034","display_name":"Digital Filter Design and Implementation","score":0.992,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9866,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/double-precision-floating-point-format","display_name":"Double-precision floating-point format","score":0.7627245},{"id":"https://openalex.org/keywords/iterative-refinement","display_name":"Iterative refinement","score":0.71254814},{"id":"https://openalex.org/keywords/single-precision-floating-point-format","display_name":"Single-precision floating-point format","score":0.70741},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5950466},{"id":"https://openalex.org/keywords/ieee-floating-point","display_name":"IEEE floating point","score":0.5905664}],"concepts":[{"id":"https://openalex.org/C35912277","wikidata":"https://www.wikidata.org/wiki/Q1243369","display_name":"Double-precision floating-point format","level":3,"score":0.7627245},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7518622},{"id":"https://openalex.org/C2779982483","wikidata":"https://www.wikidata.org/wiki/Q6094420","display_name":"Iterative refinement","level":2,"score":0.71254814},{"id":"https://openalex.org/C133095886","wikidata":"https://www.wikidata.org/wiki/Q1307173","display_name":"Single-precision floating-point format","level":3,"score":0.70741},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.6474565},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5950466},{"id":"https://openalex.org/C137231763","wikidata":"https://www.wikidata.org/wiki/Q828287","display_name":"IEEE floating point","level":3,"score":0.5905664},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5123583},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.47402576},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4521081},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.38168424},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.36444813},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.33600268},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.10403696},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec.2019.8916392","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false},{"is_oa":true,"landing_page_url":"https://www.osti.gov/biblio/1641288","pdf_url":null,"source":{"id":"https://openalex.org/S4306402487","display_name":"OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_indexed_in_scopus":false,"is_core":false,"host_organization":"https://openalex.org/I139351228","host_organization_name":"Office of Scientific and Technical Information","host_organization_lineage":["https://openalex.org/I139351228"],"host_organization_lineage_names":["Office of Scientific and Technical Information"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false}],"best_oa_location":{"is_oa":true,"landing_page_url":"https://www.osti.gov/biblio/1641288","pdf_url":null,"source":{"id":"https://openalex.org/S4306402487","display_name":"OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_indexed_in_scopus":false,"is_core":false,"host_organization":"https://openalex.org/I139351228","host_organization_name":"Office of Scientific and Technical Information","host_organization_lineage":["https://openalex.org/I139351228"],"host_organization_lineage_names":["Office of Scientific and Technical Information"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","score":0.63,"display_name":"Industry, innovation and infrastructure"}],"grants":[],"datasets":[],"versions":[],"referenced_works_count":23,"referenced_works":["https://openalex.org/W1534398032","https://openalex.org/W1841592590","https://openalex.org/W1974755392","https://openalex.org/W1975818677","https://openalex.org/W1977146902","https://openalex.org/W1995795857","https://openalex.org/W2021039675","https://openalex.org/W2028591972","https://openalex.org/W2104183282","https://openalex.org/W2169631286","https://openalex.org/W2171404670","https://openalex.org/W2395411363","https://openalex.org/W2599973081","https://openalex.org/W2606722458","https://openalex.org/W2884201348","https://openalex.org/W2895305554","https://openalex.org/W2910374893","https://openalex.org/W2963374099","https://openalex.org/W3022136271","https://openalex.org/W3150959508","https://openalex.org/W4240959780","https://openalex.org/W4242722506","https://openalex.org/W4300421229"],"related_works":["https://openalex.org/W3215589575","https://openalex.org/W3150959508","https://openalex.org/W3150370983","https://openalex.org/W2730658480","https://openalex.org/W2596474508","https://openalex.org/W2587538765","https://openalex.org/W2239119680","https://openalex.org/W2169016399","https://openalex.org/W2116803521","https://openalex.org/W1564887326"],"abstract_inverted_index":{"The":[0,174,213],"emergence":[1],"of":[2,23,33,64,78,88,119,142,166,169,186],"deep":[3],"learning":[4,12],"as":[5],"a":[6,43,62,76,120,139],"leading":[7],"computational":[8,52],"workload":[9],"for":[10,54,225],"machine":[11],"tasks":[13],"on":[14,37,209,229],"large-scale":[15],"cloud":[16],"infrastructure":[17],"installations":[18],"has":[19],"led":[20],"to":[21,46,70,84,125,182],"plethora":[22],"accelerator":[24],"hardware":[25,103,232],"releases.":[26],"However,":[27],"the":[28,34,86,111,117,164,170,187,190,218],"reduced":[29],"precision":[30],"and":[31,105,123,130,158],"range":[32,168],"floating-point":[35,98,128,211],"numbers":[36],"these":[38,48,72,230],"new":[39,140,231],"platforms":[40],"makes":[41],"it":[42,179],"non-trivial":[44],"task":[45],"leverage":[47],"unprecedented":[49],"advances":[50],"in":[51,101,189],"power":[53],"numerical":[55,219],"linear":[56,227],"algebra":[57],"operations":[58],"that":[59,80,145,178,193,222],"come":[60],"with":[61,110],"guarantee":[63],"robust":[65,226],"error":[66],"bounds.":[67],"In":[68],"order":[69],"address":[71,163],"concerns,":[73],"we":[74,95,146],"present":[75,138],"number":[77],"strategies":[79],"can":[81],"be":[82,196,202],"used":[83],"increase":[85],"accuracy":[87,188],"limited-precision":[89],"iterative":[90,206],"refinement.":[91],"By":[92],"limited":[93,167],"precision,":[94],"mean":[96],"16-bit":[97],"formats":[99],"implemented":[100],"modern":[102],"accelerators":[104],"are":[106,223],"not":[107],"necessarily":[108],"compliant":[109],"IEEE":[112,127],"half-precision":[113],"specification.":[114],"We":[115,136],"include":[116],"explanation":[118],"broader":[121],"context":[122],"connections":[124],"established":[126],"standards":[129],"existing":[131],"high-performance":[132],"computing":[133],"(HPC)":[134],"benchmarks.":[135],"also":[137,216],"formulation":[141],"LU":[143,151],"factorization":[144],"call":[147],"signed":[148],"square":[149],"root":[150],"which":[152,161],"produces":[153],"more":[154],"numerically":[155],"balanced":[156],"L":[157],"U":[159],"factors":[160],"directly":[162],"problems":[165],"low-precision":[171],"storage":[172],"formats.":[173],"experimental":[175],"results":[176],"indicate":[177],"is":[180],"possible":[181],"recover":[183],"substantial":[184],"amounts":[185],"system":[191],"solution":[192],"would":[194],"otherwise":[195],"lost.":[197],"Previously,":[198],"this":[199],"could":[200],"only":[201],"achieved":[203],"by":[204],"using":[205],"refinement":[207],"based":[208],"single-precision":[210],"arithmetic.":[212],"discussion":[214],"will":[215],"explore":[217],"stability":[220],"issues":[221],"important":[224],"solvers":[228],"platforms.":[233]},"abstract_inverted_index_v3":null,"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W2991644606","counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":1}],"updated_date":"2025-04-18T01:42:13.005365","created_date":"2019-12-05"}