{"id":"https://openalex.org/W2742800336","doi":"https://doi.org/10.26083/tuprints-00006507","title":"GPU Array Access Auto-Tuning","display_name":"GPU Array Access Auto-Tuning","publication_year":2017,"publication_date":"2017-01-01","ids":{"openalex":"https://openalex.org/W2742800336","doi":"https://doi.org/10.26083/tuprints-00006507","mag":"2742800336"},"language":"en","primary_location":{"id":"pmh:oai:tubiblio.ulb.tu-darmstadt.de:137987","doi":null,"is_oa":false,"landing_page_url":"http://tubiblio.ulb.tu-darmstadt.de/137987/","pdf_url":null,"source":{"id":"https://openalex.org/S4377196390","display_name":"TUbilio (Technical University of Darmstadt)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Dissertation"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://tuprints.ulb.tu-darmstadt.de/6507/1/main.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Weber, Nicolas","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Weber, Nicolas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994999766349792,"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":0.9994999766349792,"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/T10481","display_name":"Computer Graphics and Visualization Techniques","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1704","display_name":"Computer Graphics and Computer-Aided Design"},"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/nvidia-cuda","display_name":"NVIDIA CUDA","score":0.8709999918937683},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"memory hierarchy","score":0.656000018119812},{"id":"https://openalex.org/keywords/performance-prediction","display_name":"performance prediction","score":0.7279999852180481},{"id":"https://openalex.org/keywords/performance-portability","display_name":"performance portability","score":0.8610000014305115},{"id":"https://openalex.org/keywords/data-layout-optimization","display_name":"data layout optimization","score":0.6610000133514404},{"id":"https://openalex.org/keywords/parallel-algorithms","display_name":"parallel algorithms","score":0.3659999966621399}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8775899410247803},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8474605083465576},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.7102439403533936},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6921032667160034},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.6822715401649475},{"id":"https://openalex.org/C2778514511","wikidata":"https://www.wikidata.org/wiki/Q1374194","display_name":"Programmer","level":2,"score":0.6620963215827942},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.5330504179000854},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5022022724151611},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3545895516872406},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3481740355491638},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.19285842776298523},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.17256298661231995},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.10897421836853027},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.1011236310005188}],"mesh":[],"locations_count":5,"locations":[{"id":"pmh:oai:tubiblio.ulb.tu-darmstadt.de:137987","doi":null,"is_oa":false,"landing_page_url":"http://tubiblio.ulb.tu-darmstadt.de/137987/","pdf_url":null,"source":{"id":"https://openalex.org/S4377196390","display_name":"TUbilio (Technical University of Darmstadt)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Dissertation"},{"id":"pmh:oai:tuprints.ulb.tu-darmstadt.de:6507","doi":null,"is_oa":true,"landing_page_url":"https://tuprints.ulb.tu-darmstadt.de/6507/","pdf_url":"https://tuprints.ulb.tu-darmstadt.de/6507/1/main.pdf","source":{"id":"https://openalex.org/S4306401590","display_name":"Technischen Universit\u00e4t Darmstadt","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/doctoralThesis"},{"id":"pmh:oai:tuprints.ulb.tu-darmstadt.de:tuda/3607","doi":null,"is_oa":false,"landing_page_url":"https://doi.org/10.26083/tuprints-00006507","pdf_url":null,"source":{"id":"https://openalex.org/S4306401590","display_name":"Technischen Universit\u00e4t Darmstadt","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"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":"thesis"},{"id":"doi:10.26083/tuprints-00006507","doi":"https://doi.org/10.26083/tuprints-00006507","is_oa":true,"landing_page_url":"https://doi.org/10.26083/tuprints-00006507","pdf_url":null,"source":{"id":"https://openalex.org/S4306401590","display_name":"Technischen Universit\u00e4t Darmstadt","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dissertation"},{"id":"mag:2742800336","doi":null,"is_oa":false,"landing_page_url":"https://hgpu.org/?p=17428","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:tuprints.ulb.tu-darmstadt.de:6507","doi":null,"is_oa":true,"landing_page_url":"https://tuprints.ulb.tu-darmstadt.de/6507/","pdf_url":"https://tuprints.ulb.tu-darmstadt.de/6507/1/main.pdf","source":{"id":"https://openalex.org/S4306401590","display_name":"Technischen Universit\u00e4t Darmstadt","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/doctoralThesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320323384","display_name":"Technische Universit\u00e4t Darmstadt","ror":"https://ror.org/05n911h24"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2742800336.pdf","grobid_xml":"https://content.openalex.org/works/W2742800336.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2729851701","https://openalex.org/W2132147136","https://openalex.org/W2344642201","https://openalex.org/W2260757707","https://openalex.org/W3022572056","https://openalex.org/W2535171225","https://openalex.org/W3126298766","https://openalex.org/W2100115659","https://openalex.org/W4239030124","https://openalex.org/W2003911863"],"abstract_inverted_index":{"GPUs":[0,101,173],"have":[1,151],"been":[2,152,363],"used":[3],"for":[4,133,172,197,213,366],"years":[5],"in":[6,258],"compute":[7],"intensive":[8],"applications.":[9,216],"Their":[10],"massive":[11],"parallel":[12,35],"processing":[13],"capabilities":[14],"can":[15,275],"speedup":[16,23],"calculations":[17],"significantly.":[18],"However,":[19],"to":[20,27,43,119,130,179,219,225,229,252,268,301,319,333,354],"leverage":[21],"this":[22,160,201],"it":[24,115,128,289,330,351],"is":[25,53,78,139,266,317,331,352],"necessary":[26],"rethink":[28],"and":[29,57,71,91,105,245,274,344,346],"develop":[30],"new":[31,135],"algorithms":[32,38,52,90,299],"that":[33,84,138,154,208,361],"allow":[34],"processing.":[36],"These":[37],"are":[39],"only":[40],"one":[41],"piece":[42],"achieve":[44,220,320],"high":[45],"performance.":[46],"Nearly":[47],"as":[48,50,65],"important":[49],"suitable":[51],"the":[54,58,89,92,97,112,121,125,146,163,169,180,184,191,205,227,305,356,367,376],"actual":[55],"implementation":[56],"usage":[59],"of":[60,88,100,262,293,313,358],"special":[61],"hardware":[62,187,308],"features":[63],"such":[64],"intra-warp":[66],"communication,":[67],"shared":[68],"memory,":[69],"caches,":[70],"memory":[72,181,192],"access":[73,212],"patterns.":[74],"Optimizing":[75],"these":[76,156,221],"factors":[77],"usually":[79],"a":[80,242,246,259,291,311],"time":[81],"consuming":[82],"task":[83],"requires":[85],"deep":[86],"understanding":[87],"underlying":[93,147],"hardware.":[94],"Unlike":[95],"CPUs,":[96],"internal":[98],"structure":[99],"has":[102,129,224,362],"changed":[103],"significantly":[104],"will":[106],"likely":[107],"change":[108],"even":[109,323],"more":[110],"over":[111,183],"years.":[113],"Therefore":[114],"does":[116],"not":[117],"suffice":[118],"optimize":[120],"code":[122,144,360],"once":[123],"during":[124],"development,":[126],"but":[127],"be":[131],"optimized":[132,365],"each":[134],"GPU":[136,339,369],"generation":[137],"released.":[140],"To":[141,283],"efficiently":[142],"(re-)optimize":[143],"towards":[145],"hardware,":[148],"auto-tuning":[149],"tools":[150],"developed":[153],"perform":[155],"optimizations":[157],"automatically,":[158],"taking":[159],"burden":[161],"from":[162,297],"programmer.":[164],"In":[165,200,217,348],"particular,":[166],"NVIDIA":[167,214],"--":[168,175],"leading":[170],"manufacturer":[171],"today":[174],"applied":[176],"significant":[177],"changes":[178],"hierarchy":[182,193],"last":[185,306],"four":[186,307],"generations.":[188,347],"This":[189,250],"makes":[190],"an":[194,198],"attractive":[195],"objective":[196],"auto-tuner.":[199],"thesis":[202],"we":[203,287],"introduce":[204],"MATOG":[206,223,265,316],"auto-tuner":[207],"automatically":[209,269],"optimizes":[210],"array":[211],"CUDA":[215],"order":[218],"optimizations,":[222],"analyze":[226],"application":[228,272],"determine":[230],"optimal":[231,255],"parameter":[232,256,279],"values.":[233],"The":[234],"analysis":[235],"relies":[236],"on":[237,290,304],"empirical":[238],"profiling":[239],"combined":[240],"with":[241,310],"prediction":[243],"method":[244],"data":[247,373],"post-processing":[248],"step.":[249],"allows":[251],"find":[253],"nearly":[254],"values":[257],"minimal":[260],"amount":[261],"time.":[263],"Further,":[264,329],"able":[267,318,332,353],"detect":[270],"varying":[271],"workloads":[273],"apply":[276],"different":[277,294,338],"optimization":[278],"settings":[280],"at":[281],"runtime.":[282],"show":[284],"MATOG's":[285],"capabilities,":[286],"evaluated":[288],"variety":[292],"applications,":[295],"ranging":[296],"simple":[298],"up":[300],"complex":[302],"applications":[303],"generations,":[309],"total":[312],"14":[314],"GPUs.":[315],"equal":[321],"or":[322],"better":[324],"performance":[325,335,357],"than":[326],"hand-optimized":[327],"code.":[328],"provide":[334],"portability":[336],"across":[337],"types":[340],"(low-,":[341],"mid-,":[342],"high-end":[343],"HPC)":[345],"some":[349],"cases":[350],"exceed":[355],"hand-crafted":[359],"specifically":[364],"tested":[368],"by":[370],"dynamically":[371],"changing":[372],"layouts":[374],"throughout":[375],"execution.":[377]},"counts_by_year":[],"updated_date":"2026-11-13T13:42:25.865203","created_date":"2017-08-17T00:00:00"}
