{"id":"https://openalex.org/W4386709581","doi":"https://doi.org/10.1145/3605573.3605577","title":"FastDimeNet++: Training DimeNet++ in 22 minutes","display_name":"FastDimeNet++: Training DimeNet++ in 22 minutes","publication_year":2023,"publication_date":"2023-08-07","ids":{"openalex":"https://openalex.org/W4386709581","doi":"https://doi.org/10.1145/3605573.3605577"},"language":"en","primary_location":{"id":"doi:10.1145/3605573.3605577","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605577","pdf_url":null,"source":null,"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 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3605573.3605577","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Feiwen Zhu","orcid":"https://orcid.org/0000-0001-9397-4274"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feiwen Zhu","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0000-0001-9397-4274","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059710857","display_name":"Michal Futrega","orcid":"https://orcid.org/0000-0002-4898-8877"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michal Futrega","raw_affiliation_strings":["NVIDIA, Poland"],"raw_orcid":"https://orcid.org/0000-0002-4898-8877","affiliations":[{"raw_affiliation_string":"NVIDIA, Poland","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100657207","display_name":"Han Bao","orcid":"https://orcid.org/0000-0003-3490-0974"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han Bao","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0000-0003-3490-0974","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043943260","display_name":"Sukru Burc Eryilmaz","orcid":"https://orcid.org/0000-0002-6504-0121"},"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":"Sukru Burc Eryilmaz","raw_affiliation_strings":["NVIDIA, United States of America"],"raw_orcid":"https://orcid.org/0000-0002-6504-0121","affiliations":[{"raw_affiliation_string":"NVIDIA, United States of America","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102753066","display_name":"Fei Kong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fei Kong","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0009-0008-3960-8714","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102901210","display_name":"Kefeng Duan","orcid":"https://orcid.org/0009-0001-6731-3349"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kefeng Duan","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0009-0001-6731-3349","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092863032","display_name":"Xinnian Zheng","orcid":"https://orcid.org/0009-0003-5886-5410"},"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":"Xinnian Zheng","raw_affiliation_strings":["NVIDIA, United States of America"],"raw_orcid":"https://orcid.org/0009-0003-5886-5410","affiliations":[{"raw_affiliation_string":"NVIDIA, United States of America","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092863033","display_name":"Nimrod Angel","orcid":"https://orcid.org/0009-0006-1992-9540"},"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":"Nimrod Angel","raw_affiliation_strings":["NVIDIA, United States of America"],"raw_orcid":"https://orcid.org/0009-0006-1992-9540","affiliations":[{"raw_affiliation_string":"NVIDIA, United States of America","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101855233","display_name":"Matthias Jouanneaux","orcid":"https://orcid.org/0000-0001-9684-7246"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matthias Jouanneaux","raw_affiliation_strings":["NVIDIA, Germany"],"raw_orcid":"https://orcid.org/0000-0001-9684-7246","affiliations":[{"raw_affiliation_string":"NVIDIA, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092863030","display_name":"Maxmilian Stadler","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maxmilian Stadler","raw_affiliation_strings":["NVIDIA, Germany"],"raw_orcid":"https://orcid.org/0009-0000-6822-6047","affiliations":[{"raw_affiliation_string":"NVIDIA, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007356288","display_name":"Micha\u0142 Marcinkiewicz","orcid":"https://orcid.org/0000-0002-1316-3293"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michal Marcinkiewicz","raw_affiliation_strings":["NVIDIA, Poland"],"raw_orcid":"https://orcid.org/0000-0002-1316-3293","affiliations":[{"raw_affiliation_string":"NVIDIA, Poland","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032681704","display_name":"Fung Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fung Xie","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0009-0004-1077-8745","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071821498","display_name":"June Yang","orcid":"https://orcid.org/0009-0008-3059-7027"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"June Yang","raw_affiliation_strings":["NVIDIA, China"],"raw_orcid":"https://orcid.org/0009-0008-3059-7027","affiliations":[{"raw_affiliation_string":"NVIDIA, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008915989","display_name":"Michael Andersch","orcid":"https://orcid.org/0009-0004-5778-4480"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael Andersch","raw_affiliation_strings":["NVIDIA, Germany"],"raw_orcid":"https://orcid.org/0009-0004-5778-4480","affiliations":[{"raw_affiliation_string":"NVIDIA, Germany","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":14,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.4461,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.54395653,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"274","last_page":"284"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.992900013923645,"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/T11775","display_name":"COVID-19 diagnosis using AI","score":0.9860000014305115,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5947981476783752},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5866652727127075},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.06462401151657104}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5947981476783752},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5866652727127075},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.06462401151657104},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3605573.3605577","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605577","pdf_url":null,"source":null,"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 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3605573.3605577","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605577","pdf_url":null,"source":null,"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 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W179875071","https://openalex.org/W1598630409","https://openalex.org/W2040182267","https://openalex.org/W2139271038","https://openalex.org/W2155047849","https://openalex.org/W2778051509","https://openalex.org/W2923693308","https://openalex.org/W3008591352","https://openalex.org/W3093999435","https://openalex.org/W3104644561","https://openalex.org/W3129238601","https://openalex.org/W3152893301","https://openalex.org/W4200184446"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2376932109"],"abstract_inverted_index":{"Recently,":[0],"graph":[1,204],"neural":[2],"network":[3],"(GNN)":[4],"has":[5,47],"shown":[6],"significant":[7,83,274],"strength":[8],"in":[9,59,63,141,158],"predicting":[10],"the":[11,20,77,113,131,146,150,153,161,244,248,264],"quantum":[12,41],"mechanical":[13,42],"properties":[14,43],"of":[15,26,32,44,79,88,149,160,168,192,223,231,257,277],"molecules.":[16],"Based":[17],"on":[18,93,227],"GNN,":[19],"DimeNet++":[21,80,89,250],"leverages":[22],"both":[23],"distance":[24],"information":[25,31],"atomic":[27,33],"pairs":[28],"and":[29,46,67,185,206,210],"angle":[30],"triplets":[34],"via":[35],"message":[36],"passing":[37],"mechanism":[38],"to":[39,137,237],"predict":[40],"molecules":[45],"achieved":[48,188],"state-of-the-art":[49],"results.":[50],"However,":[51,119],"there":[52],"are":[53],"more":[54],"than":[55],"10":[56],"thousand":[57],"operators":[58],"DimeNet++,":[60],"which":[61,110,177],"results":[62,157],"low":[64,182],"GPU":[65,180,220],"utilization":[66,221],"large":[68],"CPU":[69,183],"launch":[70],"overhead.":[71],"The":[72,85],"extensive":[73,186],"time":[74,105,256],"taken":[75],"for":[76,102,263],"training":[78,86,104,255],"is":[81],"a":[82,94,142,173,190,198,219,228,253,273],"drawback.":[84],"period":[87],"exceeds":[90],"one":[91],"month":[92],"single":[95],"NVIDIA":[96],"A100":[97],"GPU.":[98,118],"A":[99],"common":[100],"method":[101],"reducing":[103,145],"involves":[106],"employing":[107],"data":[108],"parallelism,":[109],"equally":[111],"distributes":[112],"global":[114],"batch":[115],"across":[116],"each":[117],"data-parallel":[120],"task":[121],"partitioning,":[122],"by":[123],"default,":[124],"does":[125],"not":[126],"consider":[127],"load":[128,134],"imbalance":[129,135],"within":[130],"batch.":[132],"This":[133],"leads":[136],"considerable":[138],"synchronization":[139],"overhead":[140],"multi-GPU":[143],"setting,":[144],"overall":[147],"efficiency":[148],"parallelism.":[151],"For":[152],"strong-scaling":[154],"scenario,":[155],"it":[156],"32%":[159],"compute":[162],"resource":[163],"being":[164],"wasted.":[165],"In":[166,243],"light":[167],"these":[169],"observations,":[170],"we":[171,234],"propose":[172],"novel":[174],"approach,":[175],"FastDimeNet++,":[176],"delivers":[178],"high":[179],"utilization,":[181],"overhead,":[184],"scalability,":[187],"through":[189],"series":[191],"optimization":[193],"strategies.":[194],"These":[195],"include":[196],"(i)":[197],"communication-free":[199],"load-balancing":[200],"sampler,":[201],"(ii)":[202],"computation":[203],"reconstruction,":[205],"(iii)":[207],"kernel":[208],"fusion":[209],"redundancy":[211],"bypass.":[212],"Our":[213],"experiments":[214],"demonstrate":[215],"that":[216],"FastDimeNet++":[217,236,261],"achieves":[218],"rate":[222],"approximately":[224],"88%":[225],"based":[226],"mini-batch":[229],"size":[230],"4.":[232],"Furthermore,":[233],"scale":[235],"512":[238],"GPUs,":[239],"reaching":[240],"2.8":[241],"PetaFLOPS.":[242],"MLPerf":[245,265],"HPC":[246,266],"V1.0,":[247],"winning":[249],"submission":[251],"required":[252,268],"total":[254],"111.86":[258],"minutes,":[259,271],"whereas":[260],"introduced":[262],"V2.0":[267],"just":[269],"21.93":[270],"demonstrating":[272],"performance":[275],"improvement":[276],"over":[278],"5":[279],"\u00d7.":[280]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":3}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}