{"id":"https://openalex.org/W6903545278","doi":"https://doi.org/10.1184/r1/6476048","title":"On Model Parallelization and Scheduling Strategies for Distributed Machine Learning","display_name":"On Model Parallelization and Scheduling Strategies for Distributed Machine Learning","publication_year":2018,"publication_date":"2018-06-29","ids":{"openalex":"https://openalex.org/W6903545278","doi":"https://doi.org/10.1184/r1/6476048","mag":"2113547287"},"language":"en","primary_location":{"id":"pmh:oai:figshare.com:article/6476048","doi":null,"is_oa":true,"landing_page_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","pdf_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"},"volume":null,"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102015904","display_name":"Seunghak Lee","orcid":"https://orcid.org/0000-0002-8030-7138","observed_orcids":["https://orcid.org/0000-0002-8030-7138"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Seunghak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113847826","display_name":"Jin Kyu Kim","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Jin Kyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zheng, Xun","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Xun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012361506","display_name":"Qirong Ho","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ho, Qirong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041431830","display_name":"Garth A. Gibson","orcid":"https://orcid.org/0000-0002-6656-7080","observed_orcids":["https://orcid.org/0000-0002-6656-7080"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gibson, Garth","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"P Xing, Eric","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"P Xing, Eric","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":14.0584,"has_fulltext":true,"cited_by_count":87,"citation_normalized_percentile":{"value":0.99194225,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":100},"biblio":{"volume":"27","issue":null,"first_page":"2834","last_page":"2842"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.5839999914169312,"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":0.5839999914169312,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.2013999968767166,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.09300000220537186,"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/model-parallelism","display_name":"model parallelism","score":0.9390000104904175},{"id":"https://openalex.org/keywords/distributed-learning","display_name":"distributed learning","score":0.9549999833106995},{"id":"https://openalex.org/keywords/task-scheduling","display_name":"task scheduling","score":0.6700000166893005},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"data parallelism","score":0.7160000205039978},{"id":"https://openalex.org/keywords/memory-efficiency","display_name":"memory efficiency","score":0.7170000076293945},{"id":"https://openalex.org/keywords/topic-modeling","display_name":"topic modeling","score":0.5979999899864197},{"id":"https://openalex.org/keywords/matrix-factorization","display_name":"matrix factorization","score":1.0},{"id":"https://openalex.org/keywords/lasso","display_name":"LASSO","score":0.8859999775886536}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8601999878883362},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6074000000953674},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5246999859809875},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.5115000009536743},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4699999988079071},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.42489999532699585},{"id":"https://openalex.org/C95203288","wikidata":"https://www.wikidata.org/wiki/Q221682","display_name":"Semaphore","level":2,"score":0.42170000076293945},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.4174000024795532},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.415800005197525},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.3824000060558319},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.36809998750686646},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.36390000581741333},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3393000066280365},{"id":"https://openalex.org/C56348326","wikidata":"https://www.wikidata.org/wiki/Q1047554","display_name":"Mutual exclusion","level":2,"score":0.33489999175071716},{"id":"https://openalex.org/C91481028","wikidata":"https://www.wikidata.org/wiki/Q1054686","display_name":"Distributed memory","level":3,"score":0.3319999873638153},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3215999901294708},{"id":"https://openalex.org/C120373497","wikidata":"https://www.wikidata.org/wiki/Q1087987","display_name":"Parallel algorithm","level":2,"score":0.32019999623298645},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.3093000054359436},{"id":"https://openalex.org/C70061542","wikidata":"https://www.wikidata.org/wiki/Q989016","display_name":"Distributed database","level":2,"score":0.30790001153945923},{"id":"https://openalex.org/C130120984","wikidata":"https://www.wikidata.org/wiki/Q2835898","display_name":"Distributed algorithm","level":2,"score":0.30559998750686646},{"id":"https://openalex.org/C31689143","wikidata":"https://www.wikidata.org/wiki/Q733809","display_name":"Fair-share scheduling","level":3,"score":0.296099990606308},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.29190000891685486},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.29089999198913574},{"id":"https://openalex.org/C107568181","wikidata":"https://www.wikidata.org/wiki/Q5319000","display_name":"Dynamic priority scheduling","level":3,"score":0.28949999809265137},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C156891508","wikidata":"https://www.wikidata.org/wiki/Q1004114","display_name":"Bulk synchronous parallel","level":3,"score":0.266400009393692},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C2984822820","wikidata":"https://www.wikidata.org/wiki/Q1123036","display_name":"Processor scheduling","level":3,"score":0.25459998846054077},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.25200000405311584}],"mesh":[],"locations_count":6,"locations":[{"id":"pmh:oai:figshare.com:article/6476048","doi":null,"is_oa":true,"landing_page_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","pdf_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.696.6846","doi":null,"is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.696.6846","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.pdl.cmu.edu/PDL-FTP/BigLearning/NIPS2014modelparallel.pdf","raw_type":"text"},{"id":"pmh:oai:repository.cmu.edu:machine_learning-1141","doi":null,"is_oa":false,"landing_page_url":"http://repository.cmu.edu/cgi/viewcontent.cgi?article=1141&context=machine_learning","pdf_url":null,"source":{"id":"https://openalex.org/S7407050927","display_name":"KiltHub Repository","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I74973139","host_organization_name":"Carnegie Mellon University","host_organization_lineage":["https://openalex.org/I74973139"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Machine Learning Department","raw_type":"text"},{"id":"doi:10.1184/r1/6476048","doi":"https://doi.org/10.1184/r1/6476048","is_oa":true,"landing_page_url":"https://doi.org/10.1184/r1/6476048","pdf_url":null,"source":{"id":"https://openalex.org/S7407050927","display_name":"KiltHub Repository","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I74973139","host_organization_name":"Carnegie Mellon University","host_organization_lineage":["https://openalex.org/I74973139"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"JournalArticle"},{"id":"doi:10.1184/r1/6476048.v1","doi":"https://doi.org/10.1184/r1/6476048.v1","is_oa":true,"landing_page_url":"https://doi.org/10.1184/r1/6476048.v1","pdf_url":null,"source":{"id":"https://openalex.org/S7407050927","display_name":"KiltHub Repository","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I74973139","host_organization_name":"Carnegie Mellon University","host_organization_lineage":["https://openalex.org/I74973139"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"JournalArticle"},{"id":"mag:2113547287","doi":null,"is_oa":false,"landing_page_url":"https://papers.nips.cc/paper/5598-on-model-parallelization-and-scheduling-strategies-for-distributed-machine-learning.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"Neural Information Processing Systems","raw_type":null}],"best_oa_location":{"id":"pmh:oai:figshare.com:article/6476048","doi":null,"is_oa":true,"landing_page_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","pdf_url":"https://figshare.com/articles/journal_contribution/On_Model_Parallelization_and_Scheduling_Strategies_for_Distributed_Machine_Learning/6476048","source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G8496933687","display_name":"Collaborative Research: PRObE - The NSF Parallel Reconfigurable Observational Environment for Data Intensive Super-Computing and High End Computing","funder_award_id":"1042543","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8533660586","display_name":"BIGDATA: F: DKA: Collaborative Research: Theory and Algorithms for Parallel Probabilistic Inference with Big Data, via Big Model, in Realistic Distributed Computing Environments","funder_award_id":"1447676","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8790040577","display_name":null,"funder_award_id":"FA8750-12-2-0324","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320952","display_name":"International Science and Technology Center","ror":"https://ror.org/03fn1w943"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W6903545278.pdf","grobid_xml":"https://content.openalex.org/works/W6903545278.grobid-xml"},"referenced_works_count":30,"referenced_works":["https://openalex.org/W78077100","https://openalex.org/W1511814458","https://openalex.org/W1528469369","https://openalex.org/W1546359014","https://openalex.org/W1880262756","https://openalex.org/W1980147176","https://openalex.org/W1981231959","https://openalex.org/W2001082470","https://openalex.org/W2020098476","https://openalex.org/W2056636001","https://openalex.org/W2083842231","https://openalex.org/W2096544401","https://openalex.org/W2114643899","https://openalex.org/W2116137244","https://openalex.org/W2118047185","https://openalex.org/W2120480077","https://openalex.org/W2132737349","https://openalex.org/W2133233009","https://openalex.org/W2138996412","https://openalex.org/W2166706236","https://openalex.org/W2168231600","https://openalex.org/W2173213060","https://openalex.org/W2184628147","https://openalex.org/W2189465200","https://openalex.org/W2253807446","https://openalex.org/W2341535507","https://openalex.org/W2395355800","https://openalex.org/W2962727278","https://openalex.org/W2963707635","https://openalex.org/W6959825462"],"related_works":["https://openalex.org/W2335835108","https://openalex.org/W2182420680","https://openalex.org/W1497213868","https://openalex.org/W2132737349","https://openalex.org/W2119391823","https://openalex.org/W2926938941","https://openalex.org/W2415005847","https://openalex.org/W7095197347","https://openalex.org/W6922431471","https://openalex.org/W3193789877"],"abstract_inverted_index":{"Distributed":[0],"machine":[1],"learning":[2],"has":[3,43],"typically":[4],"been":[5],"approached":[6],"from":[7],"a":[8,101,108,127],"data":[9,14,28,82],"parallel":[10,56],"perspective,":[11],"where":[12,61],"big":[13],"are":[15,67],"partitioned":[16,68],"to":[17,34,50,69,135],"multiple":[18],"workers":[19,71],"and":[20,53,72,81,93,117,133,141,166],"an":[21,64],"algorithm":[22],"is":[23,48],"executed":[24],"concurrently":[25],"over":[26],"different":[27,70,86],"subsets":[29],"under":[30],"various":[31],"synchronization":[32],"schemes":[33],"ensure":[35,51],"speed-up":[36],"and/or":[37],"correctness.":[38],"A":[39],"sibling":[40],"problem":[41],"that":[42,79,106],"received":[44],"relatively":[45],"less":[46],"attention":[47],"how":[49],"efficient":[52],"correct":[54],"model":[55,80],"execution":[57],"of":[58,63,122,145,152],"ML":[59,65,123],"algorithms,":[60],"parameters":[62],"program":[66],"undergone":[73],"concurrent":[74],"iterative":[75],"updates.":[76],"We":[77,148],"argue":[78],"parallelisms":[83],"impose":[84],"rather":[85],"challenges":[87],"for":[88,103,111,161],"system":[89,102],"design,":[90],"algorithmic":[91],"adjustment,":[92],"theoretical":[94],"analysis.":[95],"In":[96],"this":[97],"paper,":[98],"we":[99],"develop":[100],"model-parallelism,":[104],"STRADS,":[105],"provides":[107],"programming":[109],"abstraction":[110],"scheduling":[112,131],"parameter":[113],"updates":[114],"by":[115],"discovering":[116],"leveraging":[118],"changing":[119],"structural":[120],"properties":[121],"programs.":[124],"STRADS":[125,157],"enables":[126],"flexible":[128],"tradeoff":[129],"between":[130],"efficiency":[132,144],"fidelity":[134],"intrinsic":[136],"dependencies":[137],"within":[138],"the":[139,150],"models,":[140],"improves":[142],"memory":[143],"distributed":[146],"ML.":[147],"demonstrate":[149],"efficacy":[151],"model-parallel":[153],"algorithms":[154],"implemented":[155],"on":[156],"versus":[158],"popular":[159],"implementations":[160],"topic":[162],"modeling,":[163],"matrix":[164],"factorization,":[165],"Lasso.":[167]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":14},{"year":2020,"cited_by_count":10},{"year":2019,"cited_by_count":18},{"year":2018,"cited_by_count":9},{"year":2017,"cited_by_count":8},{"year":2016,"cited_by_count":7},{"year":2015,"cited_by_count":3},{"year":2014,"cited_by_count":3}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2025-10-10T00:00:00"}