{"id":"https://openalex.org/W7167913414","doi":"https://doi.org/10.1145/3801488.3806377","title":"On the Limits of Performance Portability in Directive-Based GPU Programming","display_name":"On the Limits of Performance Portability in Directive-Based GPU Programming","publication_year":2026,"publication_date":"2026-05-19","ids":{"openalex":"https://openalex.org/W7167913414","doi":"https://doi.org/10.1145/3801488.3806377"},"language":null,"primary_location":{"id":"doi:10.1145/3801488.3806377","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801488.3806377","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers: Workshops and Special Sessions","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3801488.3806377","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103081063","display_name":"Alessandro B. Romeo","orcid":"https://orcid.org/0009-0001-7159-3486"},"institutions":[{"id":"https://openalex.org/I4210119786","display_name":"Cineca","ror":"https://ror.org/02f013h18","country_code":"IT","type":"other","lineage":["https://openalex.org/I4210119786"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alessandro Romeo","raw_affiliation_strings":["SuperComputing Applications and Innovation Department - CINECA, Casalecchio di Reno, Bologna, Italy"],"raw_orcid":"https://orcid.org/0009-0001-7159-3486","affiliations":[{"raw_affiliation_string":"SuperComputing Applications and Innovation Department - CINECA, Casalecchio di Reno, Bologna, Italy","institution_ids":["https://openalex.org/I4210119786"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006786147","display_name":"Nitin Shukla","orcid":"https://orcid.org/0000-0002-2110-1439"},"institutions":[{"id":"https://openalex.org/I4210119786","display_name":"Cineca","ror":"https://ror.org/02f013h18","country_code":"IT","type":"other","lineage":["https://openalex.org/I4210119786"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Nitin Shukla","raw_affiliation_strings":["SuperComputing Applications and Innovation Department - CINECA, Casalecchio di Reno, Bologna, Italy"],"raw_orcid":"https://orcid.org/0000-0002-2110-1439","affiliations":[{"raw_affiliation_string":"SuperComputing Applications and Innovation Department - CINECA, Casalecchio di Reno, Bologna, Italy","institution_ids":["https://openalex.org/I4210119786"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107316353","display_name":"Stefano Truzzi","orcid":"https://orcid.org/0000-0003-3998-9373"},"institutions":[{"id":"https://openalex.org/I55143463","display_name":"University of Turin","ror":"https://ror.org/048tbm396","country_code":"IT","type":"education","lineage":["https://openalex.org/I55143463"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Stefano Truzzi","raw_affiliation_strings":["Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy"],"raw_orcid":"https://orcid.org/0000-0003-3998-9373","affiliations":[{"raw_affiliation_string":"Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy","institution_ids":["https://openalex.org/I55143463"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126319744","display_name":"Alessio Suriano","orcid":null},"institutions":[{"id":"https://openalex.org/I55143463","display_name":"University of Turin","ror":"https://ror.org/048tbm396","country_code":"IT","type":"education","lineage":["https://openalex.org/I55143463"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alessio Suriano","raw_affiliation_strings":["Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy"],"raw_orcid":"https://orcid.org/0009-0003-1889-6239","affiliations":[{"raw_affiliation_string":"Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy","institution_ids":["https://openalex.org/I55143463"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001432824","display_name":"A. Mignone","orcid":"https://orcid.org/0000-0002-8352-6635"},"institutions":[{"id":"https://openalex.org/I55143463","display_name":"University of Turin","ror":"https://ror.org/048tbm396","country_code":"IT","type":"education","lineage":["https://openalex.org/I55143463"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Andrea Mignone","raw_affiliation_strings":["Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy"],"raw_orcid":"https://orcid.org/0000-0002-8352-6635","affiliations":[{"raw_affiliation_string":"Dipartimento di Fisica, Universit\u00e0 degli studi di Torino, Torino, Italy","institution_ids":["https://openalex.org/I55143463"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"74","last_page":"81"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9598000049591064,"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":0.9598000049591064,"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":0.01679999940097332,"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/T13200","display_name":"Spacecraft and Cryogenic Technologies","score":0.002400000113993883,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/porting","display_name":"Porting","score":0.8909000158309937},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.8276000022888184},{"id":"https://openalex.org/keywords/software-portability","display_name":"Software portability","score":0.8173999786376953},{"id":"https://openalex.org/keywords/optimizing-compiler","display_name":"Optimizing compiler","score":0.52920001745224},{"id":"https://openalex.org/keywords/profiling","display_name":"Profiling (computer programming)","score":0.4503999948501587},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.3666999936103821},{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.3449999988079071},{"id":"https://openalex.org/keywords/programming-paradigm","display_name":"Programming paradigm","score":0.3361999988555908}],"concepts":[{"id":"https://openalex.org/C106251023","wikidata":"https://www.wikidata.org/wiki/Q851989","display_name":"Porting","level":3,"score":0.8909000158309937},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8547000288963318},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.8276000022888184},{"id":"https://openalex.org/C63000827","wikidata":"https://www.wikidata.org/wiki/Q3080428","display_name":"Software portability","level":2,"score":0.8173999786376953},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6190999746322632},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.52920001745224},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.4503999948501587},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3666999936103821},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.3449999988079071},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.3393000066280365},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C133162039","wikidata":"https://www.wikidata.org/wiki/Q1061077","display_name":"Code generation","level":3,"score":0.3343000113964081},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.33090001344680786},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.31839999556541443},{"id":"https://openalex.org/C2778241615","wikidata":"https://www.wikidata.org/wiki/Q83303","display_name":"Fortran","level":2,"score":0.2971000075340271},{"id":"https://openalex.org/C139571649","wikidata":"https://www.wikidata.org/wiki/Q1156793","display_name":"Program optimization","level":3,"score":0.27649998664855957},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.27549999952316284},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.2745000123977661},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.26930001378059387},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.26820001006126404},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C76752949","wikidata":"https://www.wikidata.org/wiki/Q7607499","display_name":"Stencil","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C111564260","wikidata":"https://www.wikidata.org/wiki/Q4288856","display_name":"Interprocedural optimization","level":5,"score":0.25459998846054077}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3801488.3806377","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801488.3806377","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers: Workshops and Special Sessions","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3801488.3806377","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801488.3806377","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers: Workshops and Special Sessions","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W122620595","https://openalex.org/W1483717008","https://openalex.org/W1990090459","https://openalex.org/W2002555321","https://openalex.org/W2089088376","https://openalex.org/W2180288011","https://openalex.org/W2612954225","https://openalex.org/W2956027533","https://openalex.org/W2994151208","https://openalex.org/W2998644365","https://openalex.org/W3025571086","https://openalex.org/W3043057203","https://openalex.org/W3122423129","https://openalex.org/W3125565132","https://openalex.org/W3130554079","https://openalex.org/W3153525491","https://openalex.org/W3170817192","https://openalex.org/W3187013103","https://openalex.org/W3201201825","https://openalex.org/W4234180294","https://openalex.org/W4243474409","https://openalex.org/W4250981202","https://openalex.org/W4297097382","https://openalex.org/W4362634392","https://openalex.org/W4382601326","https://openalex.org/W4386281462","https://openalex.org/W4394906154","https://openalex.org/W4396814805","https://openalex.org/W4406157308","https://openalex.org/W4409236178","https://openalex.org/W4410096111","https://openalex.org/W4412066062","https://openalex.org/W4413412047","https://openalex.org/W7083204092","https://openalex.org/W7083624263","https://openalex.org/W7125374400","https://openalex.org/W7127910095","https://openalex.org/W7131134682","https://openalex.org/W7134990026"],"related_works":[],"abstract_inverted_index":{"The":[0],"transition":[1],"of":[2,24,113,157],"scientific":[3],"applications":[4],"to":[5,40,66,99,110,118,132,154,159],"GPU-accelerated":[6],"exascale":[7],"systems":[8],"is":[9,85],"constrained":[10],"by":[11,28,116,139],"trade-offs":[12],"between":[13],"performance,":[14],"portability,":[15],"and":[16,42,51,60,122,151,186],"productivity.":[17],"This":[18],"work":[19],"evaluates":[20],"the":[21,81,91,100,129],"performance":[22,45,64,170],"portability":[23],"directive-based":[25],"GPU":[26,172],"programming":[27],"porting":[29],"gPLUTO,":[30],"a":[31,67,72],"production-grade":[32],"magnetohydrodynamics":[33],"code":[34],"for":[35,75],"astrophysical":[36],"simulations,":[37],"from":[38],"OpenACC":[39,59,103],"OpenMP,":[41],"analyzing":[43],"its":[44],"on":[46,94],"NVIDIA":[47,57,101],"A100":[48,102],"(Leonardo":[49],"Booster)":[50],"AMD":[52,95],"MI250X":[53,96],"(LUMI-G)":[54],"devices.":[55],"On":[56],"platforms,":[58],"OpenMP":[61,83],"achieve":[62],"comparable":[63],"due":[65],"shared":[68],"compiler":[69,123,184],"backend,":[70],"providing":[71],"consistent":[73],"baseline":[74],"assessing":[76],"algorithmic":[77],"efficiency.":[78],"In":[79,142],"contrast,":[80],"same":[82],"implementation":[84],"approximately":[86],"three":[87],"times":[88],"slower":[89],"at":[90],"application":[92],"level":[93],"with":[97,105],"respect":[98],"baseline,":[104],"kernel-level":[106],"slowdowns":[107,156],"reaching":[108],"up":[109,158],"an":[111],"order":[112],"magnitude,":[114],"driven":[115],"sensitivity":[117],"strided":[119],"memory-access":[120],"patterns":[121],"limitations.":[124],"Kernel-level":[125],"profiling":[126],"shows":[127],"that":[128,168],"dominant":[130],"contributors":[131],"runtime":[133],"are":[134],"memory-latency-bound":[135],"rather":[136],"than":[137],"limited":[138],"peak":[140],"bandwidth.":[141],"low-parallelism":[143],"kernels,":[144],"C++":[145],"abstraction":[146],"layers":[147],"increase":[148],"register":[149],"pressure":[150],"spilling,":[152],"leading":[153],"extreme":[155],"47":[160],"\u00d7":[161],"in":[162,183],"specific":[163],"cases.":[164],"These":[165],"results":[166],"indicate":[167],"portable":[169],"across":[171],"architectures":[173],"requires":[174],"not":[175],"only":[176],"application-level":[177],"changes":[178],"but":[179],"also":[180],"continued":[181],"advances":[182],"backends":[185],"architecture-aware":[187],"optimization":[188],"strategies.":[189]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-07-11T00:00:00"}