{"id":"https://openalex.org/W4411260261","doi":"https://doi.org/10.1145/3729262","title":"Task-Based Tensor Computations on Modern GPUs","display_name":"Task-Based Tensor Computations on Modern GPUs","publication_year":2025,"publication_date":"2025-06-10","ids":{"openalex":"https://openalex.org/W4411260261","doi":"https://doi.org/10.1145/3729262"},"language":"en","primary_location":{"id":"doi:10.1145/3729262","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3729262","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3729262","source":{"id":"https://openalex.org/S4210216081","display_name":"Proceedings of the ACM on Programming Languages","issn_l":"2475-1421","issn":["2475-1421"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Programming Languages","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3729262","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047980263","display_name":"Rohan Yadav","orcid":"https://orcid.org/0000-0003-0746-066X"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rohan Yadav","raw_affiliation_strings":["Stanford University, Stanford, USA"],"raw_orcid":"https://orcid.org/0000-0003-0746-066X","affiliations":[{"raw_affiliation_string":"Stanford University, Stanford, USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024606205","display_name":"Michael Garland","orcid":"https://orcid.org/0000-0001-6093-7602"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael Garland","raw_affiliation_strings":["NVIDIA, Santa Clara, USA"],"raw_orcid":"https://orcid.org/0000-0001-6093-7602","affiliations":[{"raw_affiliation_string":"NVIDIA, Santa Clara, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029910004","display_name":"Alexander M. Aiken","orcid":"https://orcid.org/0000-0003-3372-9036"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alex Aiken","raw_affiliation_strings":["Stanford University, Stanford, USA"],"raw_orcid":"https://orcid.org/0000-0003-3372-9036","affiliations":[{"raw_affiliation_string":"Stanford University, Stanford, USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077152489","display_name":"Michael Bauer","orcid":"https://orcid.org/0000-0001-8928-3032"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael Bauer","raw_affiliation_strings":["NVIDIA, Santa Clara, USA"],"raw_orcid":"https://orcid.org/0000-0001-8928-3032","affiliations":[{"raw_affiliation_string":"NVIDIA, Santa Clara, USA","institution_ids":["https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.5548,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.94582691,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"9","issue":"PLDI","first_page":"396","last_page":"420"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998000264167786,"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":0.9998000264167786,"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/T13650","display_name":"Computational Physics and Python Applications","score":0.9921000003814697,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T12303","display_name":"Tensor decomposition and applications","score":0.9731000065803528,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.664665937423706},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6293171048164368},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6105965375900269},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.6073932647705078},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.5648670196533203},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5083109736442566},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.17545652389526367},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.17362689971923828},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.163334459066391},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.06705087423324585}],"concepts":[{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.664665937423706},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6293171048164368},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6105965375900269},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.6073932647705078},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.5648670196533203},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5083109736442566},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.17545652389526367},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.17362689971923828},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.163334459066391},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.06705087423324585},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3729262","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3729262","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3729262","source":{"id":"https://openalex.org/S4210216081","display_name":"Proceedings of the ACM on Programming Languages","issn_l":"2475-1421","issn":["2475-1421"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Programming Languages","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3729262","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3729262","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3729262","source":{"id":"https://openalex.org/S4210216081","display_name":"Proceedings of the ACM on Programming Languages","issn_l":"2475-1421","issn":["2475-1421"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Programming Languages","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411260261.pdf","grobid_xml":"https://content.openalex.org/works/W4411260261.grobid-xml"},"referenced_works_count":28,"referenced_works":["https://openalex.org/W1984689820","https://openalex.org/W2020165759","https://openalex.org/W2049875313","https://openalex.org/W2121893797","https://openalex.org/W2126952393","https://openalex.org/W2160523651","https://openalex.org/W2171399035","https://openalex.org/W2252007067","https://openalex.org/W2294413161","https://openalex.org/W2590246587","https://openalex.org/W2954698171","https://openalex.org/W3035965352","https://openalex.org/W3047091820","https://openalex.org/W3099525000","https://openalex.org/W3099878876","https://openalex.org/W3117137355","https://openalex.org/W3122286897","https://openalex.org/W3136479147","https://openalex.org/W3145506805","https://openalex.org/W4231715478","https://openalex.org/W4240382083","https://openalex.org/W4244157805","https://openalex.org/W4248722156","https://openalex.org/W4251637954","https://openalex.org/W4280495398","https://openalex.org/W4281707342","https://openalex.org/W4327930469","https://openalex.org/W4396817063"],"related_works":["https://openalex.org/W4231775656","https://openalex.org/W2046435967","https://openalex.org/W2383646825","https://openalex.org/W2371018915","https://openalex.org/W2354191502","https://openalex.org/W1972225038","https://openalex.org/W3134658850","https://openalex.org/W2355938171","https://openalex.org/W2780079842","https://openalex.org/W1429949169"],"abstract_inverted_index":{"Domain-specific,":[0],"fixed-function":[1,24,35],"units":[2,25,36,58],"are":[3,104,117,125],"becoming":[4],"increasingly":[5],"common":[6],"in":[7,142],"modern":[8],"processors.":[9],"As":[10],"the":[11,17,80,85,128,170,179,182],"computational":[12],"demands":[13],"of":[14,22,87,107,119,172,181,192],"applications":[15],"evolve,":[16],"capabilities":[18],"and":[19,47,116,121,141,176,196],"programming":[20,63,88,97],"interfaces":[21],"these":[23,57,89],"continue":[26],"to":[27,127],"change.":[28],"NVIDIA\u2019s":[29],"Hopper":[30],"GPU":[31],"architecture":[32,153],"contains":[33],"multiple":[34],"per":[37],"compute":[38],"unit,":[39],"including":[40],"an":[41,48],"asynchronous":[42,49,81,197],"data":[43,194],"movement":[44,195],"unit":[45,52],"(TMA)":[46],"matrix":[50],"multiplication":[51],"(Tensor":[53],"Core).":[54],"Efficiently":[55],"utilizing":[56],"requires":[59],"a":[60,95,105,132,151],"fundamentally":[61],"different":[62],"style":[64],"than":[65],"previous":[66],"architectures;":[67],"programmers":[68],"must":[69],"now":[70],"develop":[71],"warp-specialized":[72],"kernels":[73],"that":[74,112,135,154,161],"orchestrate":[75],"producer":[76],"consumer":[77],"pipelines":[78],"between":[79,177],"units.":[82],"To":[83],"manage":[84],"complexity":[86],"new":[90],"architectures,":[91],"we":[92],"introduce":[93],"Cypress,":[94],"task-based":[96],"model":[98],"with":[99,164],"sequential":[100],"semantics.":[101],"Cypress":[102,123,156,167],"programs":[103,124,157,160],"set":[106],"designated":[108],"functions":[109],"called":[110],"tasks":[111,138],"operate":[113],"on":[114,174],"tensors":[115,145],"free":[118],"communication":[120],"synchronization.":[122],"bound":[126],"target":[129],"machine":[130],"through":[131],"mapping":[133],"specification":[134],"describes":[136],"where":[137],"should":[139,146],"run":[140],"which":[143],"memories":[144],"be":[147],"materialized.":[148],"We":[149],"present":[150],"compiler":[152],"lowers":[155],"into":[158],"CUDA":[159],"perform":[162],"competitively":[163],"expert-written":[165],"codes.":[166],"achieves":[168],"0.88x-1.06x":[169],"performance":[171,180],"cuBLAS":[173],"GEMM,":[175],"0.80x-0.98x":[178],"currently":[183],"best-known":[184],"Flash":[185],"Attention":[186],"implementation":[187],"while":[188],"eliminating":[189],"all":[190],"aspects":[191],"explicit":[193],"computation":[198],"from":[199],"application":[200],"code.":[201]},"counts_by_year":[{"year":2026,"cited_by_count":5}],"updated_date":"2026-08-11T07:18:39.950985","created_date":"2025-10-10T00:00:00"}