{"id":"https://openalex.org/W3039165326","doi":"https://doi.org/10.1145/3392717.3392771","title":"NV-group","display_name":"NV-group","publication_year":2020,"publication_date":"2020-06-29","ids":{"openalex":"https://openalex.org/W3039165326","doi":"https://doi.org/10.1145/3392717.3392771","mag":"3039165326"},"language":"en","primary_location":{"id":"doi:10.1145/3392717.3392771","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3392717.3392771","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Supercomputing","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076405830","display_name":"Ching-Hsiang Chu","orcid":"https://orcid.org/0000-0002-6752-3135"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ching-Hsiang Chu","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057608531","display_name":"Pouya Kousha","orcid":"https://orcid.org/0009-0004-7507-0940"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pouya Kousha","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004330728","display_name":"Ammar Ahmad Awan","orcid":"https://orcid.org/0000-0002-6272-3760"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ammar Ahmad Awan","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017245523","display_name":"Kawthar Shafie Khorassani","orcid":"https://orcid.org/0000-0001-5856-5483"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kawthar Shafie Khorassani","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034293705","display_name":"Hari Subramoni","orcid":"https://orcid.org/0000-0002-1200-2754"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hari Subramoni","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024879682","display_name":"Dhabaleswar K. Panda","orcid":"https://orcid.org/0000-0002-0356-1781"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dhabaleswar K. (D K) Panda","raw_affiliation_strings":["The Ohio State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":2.3026,"has_fulltext":false,"cited_by_count":39,"citation_normalized_percentile":{"value":0.92082177,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"12"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9973000288009644,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9972000122070312,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8757654428482056},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8565331697463989},{"id":"https://openalex.org/keywords/graphics-processing-unit","display_name":"Graphics processing unit","score":0.6689674854278564},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.61655193567276},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.458236962556839},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4511517584323883},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.4447939097881317},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.42311811447143555},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.41651758551597595},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.38793373107910156},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.15320256352424622},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1381666362285614}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8757654428482056},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8565331697463989},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.6689674854278564},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.61655193567276},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.458236962556839},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4511517584323883},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.4447939097881317},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.42311811447143555},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.41651758551597595},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.38793373107910156},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.15320256352424622},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1381666362285614}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3392717.3392771","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3392717.3392771","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Supercomputing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4000000059604645,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[{"id":"https://openalex.org/G7474096879","display_name":null,"funder_award_id":"1931537,1450440,1664137,1818253","funder_id":"https://openalex.org/F4320315254","funder_display_name":"Innovative Research Group Project of the National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320315254","display_name":"Innovative Research Group Project of the National Natural Science Foundation of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W1572016165","https://openalex.org/W1964981582","https://openalex.org/W2057332538","https://openalex.org/W2131613942","https://openalex.org/W2134427337","https://openalex.org/W2160054705","https://openalex.org/W2488497244","https://openalex.org/W2580688187","https://openalex.org/W2612387305","https://openalex.org/W2808274640","https://openalex.org/W2904180680","https://openalex.org/W2908125138","https://openalex.org/W2951964584","https://openalex.org/W2972087877","https://openalex.org/W2976292093","https://openalex.org/W2981114289","https://openalex.org/W2985247698","https://openalex.org/W3104002337","https://openalex.org/W3151433020","https://openalex.org/W4245962127","https://openalex.org/W4367148982"],"related_works":["https://openalex.org/W2983282793","https://openalex.org/W1973046741","https://openalex.org/W1963859303","https://openalex.org/W2364044215","https://openalex.org/W2389600408","https://openalex.org/W240129890","https://openalex.org/W2056717482","https://openalex.org/W2030707850","https://openalex.org/W2393490604","https://openalex.org/W2375309128"],"abstract_inverted_index":{"The":[0],"advanced":[1],"fabrics":[2],"like":[3,177],"NVIDIA":[4,114],"NVLink":[5],"are":[6],"enabling":[7],"the":[8,23,45,145,150,157],"deployment":[9],"of":[10,26,85,103,144,152],"dense":[11],"Graphics":[12],"Processing":[13],"Unit":[14],"(GPU)":[15],"systems":[16,29,176],"such":[17,43],"as":[18,44,93,95],"DGX-2":[19,139,178],"and":[20,80,82,91,113,121,127,141,169,179],"Summit.":[21],"With":[22],"wide":[24],"adoption":[25],"large-scale":[27],"GPU-enabled":[28],"for":[30,72,130,165,174],"distributed":[31,73,166],"deep":[32,74],"learning":[33,75],"(DL)":[34],"training,":[35],"it":[36],"is":[37,156],"vital":[38],"to":[39,48,67,109],"design":[40],"efficient":[41],"communication":[42,81],"Allreduce":[46,70,104],"operation":[47],"achieve":[49],"near-ideal":[50,162],"speedup":[51],"at":[52],"scale.":[53],"In":[54],"this":[55,155],"paper,":[56],"we":[57,98,118],"propose":[58],"a":[59,137],"link-efficient":[60],"scheme":[61],"through":[62],"NVLink-aware":[63],"cooperative":[64],"reduction":[65],"kernels":[66],"significantly":[68],"accelerate":[69],"operations":[71],"applications.":[76],"By":[77],"overlapping":[78],"computation":[79],"maximizing":[83],"utilization":[84],"all":[86],"available":[87],"NVLinks":[88],"between":[89],"CPU":[90],"GPU,":[92],"well":[94],"among":[96],"GPUs,":[97],"demonstrate":[99,119],"1.8X":[100],"performance":[101],"improvement":[102],"on":[105,136,142],"1,536":[106],"GPUs":[107],"compared":[108],"state-of-the-art":[110],"GPU-Aware":[111],"MPI":[112],"NCCL":[115],"libraries.":[116],"Finally,":[117],"93.9%":[120],"89.7%":[122],"scaling":[123,163],"efficiency":[124,164],"(i.e.,":[125],"15X":[126],"172X":[128],"speedup)":[129],"training":[131,168],"ResNet-50":[132],"models":[133],"using":[134],"TensorFlow":[135],"16-GPU":[138],"node":[140],"192-GPUs":[143],"Summit":[146,180],"system,":[147],"respectively.":[148],"To":[149],"best":[151],"our":[153],"knowledge,":[154],"first":[158],"study":[159],"that":[160],"achieves":[161],"DL":[167],"deals":[170],"with":[171],"designs":[172],"tailored":[173],"cutting-edge":[175],"clusters.":[181]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":11},{"year":2021,"cited_by_count":9},{"year":2020,"cited_by_count":2}],"updated_date":"2026-08-22T07:34:49.880490","created_date":"2020-07-10T00:00:00"}