{"id":"https://openalex.org/W4388031374","doi":"https://doi.org/10.1145/3581784.3607062","title":"Parallel Top-K Algorithms on GPU: A Comprehensive Study and New Methods","display_name":"Parallel Top-K Algorithms on GPU: A Comprehensive Study and New Methods","alternate_titles":[],"publication_year":2023,"publication_date":"2023-11-11","ids":{"openalex":"https://openalex.org/W4388031374","doi":"https://doi.org/10.1145/3581784.3607062"},"language":"en","primary_location":{"id":"doi:10.1145/3581784.3607062","doi":"https://doi.org/10.1145/3581784.3607062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3581784.3607062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3581784.3607062","source":{"id":"https://openalex.org/S4363608557","display_name":"International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","issn_l":"2167-4337","issn":["2167-4337"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"parent_works":[],"lineage":["https://openalex.org/W4388031374"],"child_works_count":0,"type":"conference-paper","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3581784.3607062","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075095992","display_name":"Jingrong Zhang","orcid":"https://orcid.org/0000-0002-7829-7348","observed_orcids":["https://orcid.org/0000-0002-7829-7348"]},"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,"contributor_roles":[],"raw_author_name":"Jingrong Zhang","raw_affiliation_strings":["NVIDIA, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-7829-7348","affiliations":[{"raw_affiliation_string":"NVIDIA, Shanghai, China","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050692022","display_name":"Akira Naruse","orcid":"https://orcid.org/0000-0002-3140-0854","observed_orcids":["https://orcid.org/0000-0002-3140-0854"]},"institutions":[],"countries":[],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Akira Naruse","raw_affiliation_strings":["NVIDIA, Tokyo, Japan"],"raw_orcid":"https://orcid.org/0000-0002-3140-0854","affiliations":[{"raw_affiliation_string":"NVIDIA, Tokyo, Japan","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009047207","display_name":"Xipeng Li","orcid":"https://orcid.org/0009-0009-9606-1599","observed_orcids":["https://orcid.org/0009-0009-9606-1599"]},"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,"contributor_roles":[],"raw_author_name":"Xipeng Li","raw_affiliation_strings":["NVIDIA, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0009-9606-1599","affiliations":[{"raw_affiliation_string":"NVIDIA, Beijing, China","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101762937","display_name":"Yong Wang","orcid":"https://orcid.org/0009-0005-0906-8778","observed_orcids":["https://orcid.org/0009-0005-0906-8778"]},"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,"contributor_roles":[],"raw_author_name":"Yong Wang","raw_affiliation_strings":["NVIDIA, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0005-0906-8778","affiliations":[{"raw_affiliation_string":"NVIDIA, Shanghai, China","institution_ids":["https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.5367,"has_fulltext":true,"cited_by_count":40,"citation_normalized_percentile":{"value":0.97212601,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":86,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"13"},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.989300012588501,"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":0.989300012588501,"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":0.002400000113993883,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.0015999999595806003,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/gpu-computing","display_name":"GPU computing","score":0.7850000262260437},{"id":"https://openalex.org/keywords/parallel-algorithms","display_name":"parallel algorithms","score":0.8989999890327454},{"id":"https://openalex.org/keywords/cpu-gpu-communication","display_name":"CPU-GPU communication","score":0.8880000114440918},{"id":"https://openalex.org/keywords/adaptive-algorithms","display_name":"adaptive algorithms","score":0.5360000133514404},{"id":"https://openalex.org/keywords/shared-queues","display_name":"shared queues","score":0.6909999847412109},{"id":"https://openalex.org/keywords/data-distribution","display_name":"data distribution","score":0.39899998903274536}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.9514505863189697},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8326869010925293},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7246891856193542},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.49448418617248535},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.4712265729904175},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3581784.3607062","doi":"https://doi.org/10.1145/3581784.3607062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3581784.3607062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3581784.3607062","source":{"id":"https://openalex.org/S4363608557","display_name":"International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","issn_l":"2167-4337","issn":["2167-4337"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3581784.3607062","doi":"https://doi.org/10.1145/3581784.3607062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3581784.3607062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3581784.3607062","source":{"id":"https://openalex.org/S4363608557","display_name":"International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","issn_l":"2167-4337","issn":["2167-4337"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388031374.pdf","grobid_xml":"https://content.openalex.org/works/W4388031374.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1480958225","https://openalex.org/W1506582611","https://openalex.org/W2058244017","https://openalex.org/W2058942559","https://openalex.org/W2080928154","https://openalex.org/W2124509324","https://openalex.org/W2140523364","https://openalex.org/W2468923260","https://openalex.org/W2593864460","https://openalex.org/W2735481939","https://openalex.org/W2747329762","https://openalex.org/W2798445803","https://openalex.org/W2900229157","https://openalex.org/W2902572901","https://openalex.org/W2949985202","https://openalex.org/W2951465128","https://openalex.org/W2984796925","https://openalex.org/W3003611599","https://openalex.org/W3040206323","https://openalex.org/W3093933627","https://openalex.org/W3112474878","https://openalex.org/W3135737983","https://openalex.org/W3200002085","https://openalex.org/W4285582629"],"related_works":["https://openalex.org/W4399282116","https://openalex.org/W4391959512","https://openalex.org/W2798445803","https://openalex.org/W3200002085","https://openalex.org/W4393316497","https://openalex.org/W1967385114","https://openalex.org/W4403665790","https://openalex.org/W4404681381","https://openalex.org/W2990976171","https://openalex.org/W4405095625"],"abstract_inverted_index":{"The":[0],"top-K":[1,24,32,130],"problem":[2,108],"is":[3],"an":[4,48],"essential":[5],"part":[6],"of":[7,90,107],"many":[8],"important":[9],"applications":[10],"in":[11],"scientific":[12],"computing,":[13],"information":[14],"retrieval,":[15],"etc.":[16],"As":[17],"data":[18,57,70,75,111],"volume":[19],"grows":[20],"rapidly,":[21],"high-performance":[22],"parallel":[23,31,83],"algorithms":[25],"become":[26],"critical.":[27],"We":[28,93],"propose":[29],"two":[30],"algorithms,":[33],"AIR":[34,45,120],"Top-K":[35,46,121],"(Adaptive":[36],"and":[37,41,55,82,100,110,117,124,132,134],"Iteration-fused":[38],"Radix":[39],"Top-K)":[40],"GridSelect,":[42],"for":[43,103],"GPU.":[44],"employs":[47],"iteration-fused":[49],"design":[50],"to":[51,86,143],"minimize":[52],"CPU-GPU":[53],"communication":[54],"device":[56,64],"access.":[58],"Its":[59],"adaptive":[60],"strategy":[61],"eliminates":[62],"unnecessary":[63],"memory":[65],"traffic":[66],"automatically":[67],"under":[68],"various":[69],"distributions.":[71,112],"GridSelect":[72,140],"can":[73],"process":[74],"on-the-fly.":[76],"It":[77],"adopts":[78],"a":[79,104],"shared":[80],"queue":[81],"two-step":[84],"insertion":[85],"decrease":[87],"the":[88],"frequency":[89],"costly":[91],"operations.":[92],"comprehensively":[94],"compare":[95],"8":[96],"open-source":[97],"GPU":[98],"implementations":[99],"our":[101],"methods":[102],"wide":[105],"range":[106],"sizes":[109,115],"For":[113],"batch":[114],"1":[116],"100,":[118],"respectively,":[119],"shows":[122,141],"1.98--21.48\u00d7":[123],"8.01--574.78\u00d7":[125],"speedup":[126,136,145],"over":[127,137,146],"previous":[128],"radix":[129],"algorithm,":[131],"1.44--7.34\u00d7":[133],"1.38--31.91\u00d7":[135],"state-of-the-art":[138],"methods.":[139],"up":[142],"882.29\u00d7":[144],"its":[147],"baseline.":[148]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4388031374","counts_by_year":[{"year":2026,"cited_by_count":18},{"year":2025,"cited_by_count":15},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":1}],"updated_date":"2026-11-20T13:11:24.684918","created_date":"2023-11-01T00:00:00"}
