{"id":"https://openalex.org/W2963339410","doi":"https://doi.org/10.1109/padsw.2018.8644932","title":"A DAG Model of Synchronous Stochastic Gradient Descent in Distributed Deep Learning","display_name":"A DAG Model of Synchronous Stochastic Gradient Descent in Distributed Deep Learning","publication_year":2018,"publication_date":"2018-12-01","ids":{"openalex":"https://openalex.org/W2963339410","doi":"https://doi.org/10.1109/padsw.2018.8644932","mag":"2963339410"},"language":"en","primary_location":{"id":"doi:10.1109/padsw.2018.8644932","is_oa":false,"landing_page_url":"https://doi.org/10.1109/padsw.2018.8644932","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS)","raw_type":"proceedings-article"},"type":"article","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/A5016836702","display_name":"Shaohuai Shi","orcid":"https://orcid.org/0000-0002-1418-5160"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Shaohuai Shi","raw_affiliation_strings":["Department of Computer Science, Hong Kong Baptist University"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Hong Kong Baptist University","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018867983","display_name":"Qiang Wang","orcid":"https://orcid.org/0000-0002-2986-967X"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Qiang Wang","raw_affiliation_strings":["Hong Kong Baptist University, Kowloon, HK, HK"],"affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Kowloon, HK, HK","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100730785","display_name":"Xiaowen Chu","orcid":"https://orcid.org/0000-0001-9745-4372"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiaowen Chu","raw_affiliation_strings":["Department of Computer Science, Hong Kong Baptist University"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Hong Kong Baptist University","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100729678","display_name":"Bo Li","orcid":"https://orcid.org/0000-0002-3330-8103"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Bo Li","raw_affiliation_strings":["Department of Computer Science and Engineering, The Hong Kong University of Science and Technology"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, The Hong Kong University of Science and Technology","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5016836702"],"corresponding_institution_ids":["https://openalex.org/I141568987"],"apc_list":null,"apc_paid":null,"fwci":1.1701,"has_fulltext":false,"cited_by_count":20,"citation_normalized_percentile":{"value":0.84559469,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"425","last_page":"432"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"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":0.9995999932289124,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9994999766349792,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9984999895095825,"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":0.8527711629867554},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.822262167930603},{"id":"https://openalex.org/keywords/infiniband","display_name":"InfiniBand","score":0.75323486328125},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.6999222636222839},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6122652888298035},{"id":"https://openalex.org/keywords/directed-acyclic-graph","display_name":"Directed acyclic graph","score":0.5522149801254272},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5403795838356018},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.4353001117706299},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3806415796279907},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.31798920035362244},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.2750783860683441},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1488150656223297}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8527711629867554},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.822262167930603},{"id":"https://openalex.org/C2781030343","wikidata":"https://www.wikidata.org/wiki/Q922437","display_name":"InfiniBand","level":2,"score":0.75323486328125},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.6999222636222839},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6122652888298035},{"id":"https://openalex.org/C74197172","wikidata":"https://www.wikidata.org/wiki/Q1195339","display_name":"Directed acyclic graph","level":2,"score":0.5522149801254272},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5403795838356018},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4353001117706299},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3806415796279907},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.31798920035362244},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.2750783860683441},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1488150656223297},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/padsw.2018.8644932","is_oa":false,"landing_page_url":"https://doi.org/10.1109/padsw.2018.8644932","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS)","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-96482","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-96482","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"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":"Conference paper"},{"id":"pmh:oai:repository.ust.hk:1783.1-96482","is_oa":false,"landing_page_url":"http://www.scopus.com/record/display.url?eid=2-s2.0-85063348673&origin=inward","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W1667652561","https://openalex.org/W2060393849","https://openalex.org/W2083842231","https://openalex.org/W2097117768","https://openalex.org/W2108598243","https://openalex.org/W2111713407","https://openalex.org/W2113547287","https://openalex.org/W2120432001","https://openalex.org/W2163605009","https://openalex.org/W2166706236","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2287011250","https://openalex.org/W2332506150","https://openalex.org/W2336650964","https://openalex.org/W2339765813","https://openalex.org/W2513383847","https://openalex.org/W2535374105","https://openalex.org/W2580688187","https://openalex.org/W2622263826","https://openalex.org/W2734556049","https://openalex.org/W2736230459","https://openalex.org/W2740001873","https://openalex.org/W2749988060","https://openalex.org/W2759465730","https://openalex.org/W2760303966","https://openalex.org/W2919115771","https://openalex.org/W2962758826","https://openalex.org/W2963227930","https://openalex.org/W2963674387","https://openalex.org/W2964140730","https://openalex.org/W2964324519","https://openalex.org/W4293318097","https://openalex.org/W4297827580","https://openalex.org/W4302296459","https://openalex.org/W6637151318","https://openalex.org/W6665801690","https://openalex.org/W6676756733","https://openalex.org/W6684191040","https://openalex.org/W6684249991","https://openalex.org/W6686509673","https://openalex.org/W6687483927","https://openalex.org/W6702429100","https://openalex.org/W6703420464","https://openalex.org/W6739622702","https://openalex.org/W6739693220","https://openalex.org/W6743289643","https://openalex.org/W6744711479","https://openalex.org/W6745362205","https://openalex.org/W6746239207","https://openalex.org/W6765060121"],"related_works":["https://openalex.org/W2406331183","https://openalex.org/W2363988059","https://openalex.org/W2053215237","https://openalex.org/W2792231649","https://openalex.org/W1939358748","https://openalex.org/W1995426833","https://openalex.org/W3150273738","https://openalex.org/W2613555492","https://openalex.org/W2141107779","https://openalex.org/W2350246940"],"abstract_inverted_index":{"With":[0,39],"huge":[1],"amounts":[2],"of":[3,43,102,167],"training":[4,27,61,106],"data,":[5],"deep":[6,94,117],"learning":[7,95,118],"has":[8,88],"made":[9],"great":[10],"breakthroughs":[11],"in":[12,92],"many":[13],"artificial":[14],"intelligence":[15],"(AI)":[16],"applications.":[17],"However,":[18],"such":[19],"large-scale":[20],"data":[21,49,103,133,165],"sets":[22],"present":[23],"computational":[24],"challenges,":[25],"requiring":[26],"to":[28,76,177],"be":[29,157,175],"distributed":[30,79,93,116],"on":[31,58,105,113],"a":[32,55,69],"cluster":[33],"equipped":[34],"with":[35,131],"accelerators":[36],"like":[37],"GPUs.":[38],"the":[40,48,59,78,99,150,164],"fast":[41],"increase":[42],"G":[44],"PU":[45],"computing":[46],"power,":[47],"communications":[50,104],"among":[51],"GPUs":[52],"have":[53],"become":[54],"potential":[56,151],"bottleneck":[57],"overall":[60],"performance.":[62],"In":[63],"this":[64],"paper,":[65],"we":[66,108,148,162],"first":[67],"propose":[68],"general":[70],"directed":[71],"acyclic":[72],"graph":[73],"(DAG)":[74],"model":[75],"describe":[77],"synchronous":[80],"stochastic":[81],"gradient":[82],"descent":[83],"(S-SG":[84],"D)":[85],"algorithm,":[86],"which":[87,173],"been":[89],"widely":[90],"used":[91,176],"frameworks.":[96],"To":[97],"understand":[98],"practical":[100],"impact":[101],"performance,":[107],"conduct":[109],"extensive":[110],"empirical":[111],"studies":[112],"four":[114],"state-of-the-art":[115],"frameworks":[119],"(i.e.,":[120],"Caffe-MPI,":[121],"CNTK,":[122],"MXNet":[123],"and":[124,128,140,145,153],"TensorFlow)":[125],"over":[126],"multi-GPU":[127],"multi-node":[129],"environments":[130],"different":[132],"communication":[134],"techniques,":[135],"including":[136],"PCIe,":[137],"NVLink,":[138],"10GbE,":[139],"InfiniBand.":[141],"Through":[142],"both":[143],"analytical":[144],"experimental":[146,169],"studies,":[147],"identify":[149],"bottlenecks":[152],"overheads":[154],"that":[155],"could":[156,174],"further":[158],"optimized.":[159],"At":[160],"last,":[161],"make":[163],"set":[166],"our":[168],"traces":[170],"publicly":[171],"available,":[172],"support":[178],"simulation-based":[179],"studies.":[180]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":3},{"year":2019,"cited_by_count":4}],"updated_date":"2026-03-28T08:17:26.163206","created_date":"2025-10-10T00:00:00"}