{"id":"https://openalex.org/W3204802562","doi":"https://doi.org/10.1109/icdcs51616.2021.00085","title":"Harmony: A Scheduling Framework Optimized for Multiple Distributed Machine Learning Jobs","display_name":"Harmony: A Scheduling Framework Optimized for Multiple Distributed Machine Learning Jobs","publication_year":2021,"publication_date":"2021-07-01","ids":{"openalex":"https://openalex.org/W3204802562","doi":"https://doi.org/10.1109/icdcs51616.2021.00085","mag":"3204802562"},"language":"en","primary_location":{"id":"doi:10.1109/icdcs51616.2021.00085","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icdcs51616.2021.00085","pdf_url":null,"source":{"id":"https://openalex.org/S4363608224","display_name":"IEEE International Conference on Distributed Computing Systems (ICDCS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE 41st International Conference on Distributed Computing Systems (ICDCS)","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/A5039177659","display_name":"Woo-Yeon Lee","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Woo-Yeon Lee","raw_affiliation_strings":["Samsung Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101833740","display_name":"Yunseong Lee","orcid":"https://orcid.org/0000-0001-9245-2968","observed_orcids":["https://orcid.org/0000-0001-9245-2968"]},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yunseong Lee","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004695340","display_name":"Won Wook Song","orcid":"https://orcid.org/0000-0002-8530-2184","observed_orcids":["https://orcid.org/0000-0002-8530-2184"]},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Won Wook Song","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073650020","display_name":"Youngseok Yang","orcid":"https://orcid.org/0000-0001-6246-8790","observed_orcids":["https://orcid.org/0000-0001-6246-8790"]},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youngseok Yang","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100741456","display_name":"Joo Yeon Kim","orcid":"https://orcid.org/0000-0001-6657-9578","observed_orcids":["https://orcid.org/0000-0001-6657-9578"]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Joo Yeon Kim","raw_affiliation_strings":["Samsung Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5083084972","display_name":"Byung-Gon Chun","orcid":"https://orcid.org/0000-0002-9863-7186","observed_orcids":["https://orcid.org/0000-0002-9863-7186","https://orcid.org/0009-0006-2356-971X"]},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Byung-Gon Chun","raw_affiliation_strings":["FriendliAI","Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FriendliAI","institution_ids":[]},{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.8825,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.74084792,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":88,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"841","last_page":"851"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9429000020027161,"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"}},"topics":[{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9429000020027161,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.04390000179409981,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.006599999964237213,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/job-scheduling","display_name":"job scheduling","score":0.6729999780654907},{"id":"https://openalex.org/keywords/distributed-learning","display_name":"distributed learning","score":0.8790000081062317},{"id":"https://openalex.org/keywords/parameter-server","display_name":"parameter server","score":0.8149999976158142},{"id":"https://openalex.org/keywords/cluster-resource-utilization","display_name":"cluster resource utilization","score":0.7400000095367432},{"id":"https://openalex.org/keywords/resource-contention","display_name":"resource contention","score":0.6880000233650208},{"id":"https://openalex.org/keywords/memory-pressure","display_name":"memory pressure","score":0.7279999852180481},{"id":"https://openalex.org/keywords/iterative-training","display_name":"iterative training","score":0.5249999761581421}],"concepts":[{"id":"https://openalex.org/C2776453491","wikidata":"https://www.wikidata.org/wiki/Q5659234","display_name":"Harmony (color)","level":2,"score":0.834023118019104},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.773246169090271},{"id":"https://openalex.org/C33099171","wikidata":"https://www.wikidata.org/wiki/Q26208718","display_name":"Harmony search","level":2,"score":0.7557665705680847},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.6730409860610962},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6064813733100891},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5905131101608276},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.47030967473983765},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.4372996687889099},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3464013934135437},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3419506847858429},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3159119486808777},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.23498836159706116},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.19796296954154968},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icdcs51616.2021.00085","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icdcs51616.2021.00085","pdf_url":null,"source":{"id":"https://openalex.org/S4363608224","display_name":"IEEE International Conference on Distributed Computing Systems (ICDCS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE 41st International Conference on Distributed Computing Systems (ICDCS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"display_name":"Decent work and economic growth","score":0.6499999761581421,"id":"https://metadata.un.org/sdg/8"}],"study_designs":[],"awards":[{"id":"https://openalex.org/G944557987","display_name":null,"funder_award_id":"20\u20130139","funder_id":"https://openalex.org/F4320315121","funder_display_name":"Samsung Advanced Institute of Technology"}],"funders":[{"id":"https://openalex.org/F4320315121","display_name":"Samsung Advanced Institute of Technology","ror":"https://ror.org/04axnyp10"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":54,"referenced_works":["https://openalex.org/W1442374986","https://openalex.org/W1528469369","https://openalex.org/W1652793671","https://openalex.org/W1833917188","https://openalex.org/W1969805974","https://openalex.org/W1982063824","https://openalex.org/W2060393849","https://openalex.org/W2083842231","https://openalex.org/W2127941149","https://openalex.org/W2132737349","https://openalex.org/W2336650964","https://openalex.org/W2339765813","https://openalex.org/W2341535507","https://openalex.org/W2402144811","https://openalex.org/W2528415359","https://openalex.org/W2535724050","https://openalex.org/W2537369414","https://openalex.org/W2605643718","https://openalex.org/W2609354095","https://openalex.org/W2612026221","https://openalex.org/W2762551481","https://openalex.org/W2797862701","https://openalex.org/W2798515322","https://openalex.org/W2807992755","https://openalex.org/W2899071864","https://openalex.org/W2906007643","https://openalex.org/W2919594608","https://openalex.org/W2945003059","https://openalex.org/W2948349252","https://openalex.org/W2950445386","https://openalex.org/W2953384591","https://openalex.org/W2962684017","https://openalex.org/W2962725887","https://openalex.org/W2962885409","https://openalex.org/W2963403751","https://openalex.org/W2982612373","https://openalex.org/W2984509121","https://openalex.org/W3102385572","https://openalex.org/W4234138276","https://openalex.org/W6628377381","https://openalex.org/W6636806777","https://openalex.org/W6638533276","https://openalex.org/W6665801690","https://openalex.org/W6679154944","https://openalex.org/W6679393576","https://openalex.org/W6703420464","https://openalex.org/W6703949738","https://openalex.org/W6713134421","https://openalex.org/W6727576777","https://openalex.org/W6728865109","https://openalex.org/W6752397060","https://openalex.org/W6756009870","https://openalex.org/W6758283263","https://openalex.org/W6759814162"],"related_works":["https://openalex.org/W2163159199","https://openalex.org/W101238908","https://openalex.org/W2139982839","https://openalex.org/W1984267569","https://openalex.org/W2092966558","https://openalex.org/W1956651153","https://openalex.org/W2347561926","https://openalex.org/W2327547880","https://openalex.org/W3153517983","https://openalex.org/W2842402481"],"abstract_inverted_index":{"We":[0],"introduce":[1],"Harmony,":[2],"a":[3,22,58,86],"new":[4],"scheduling":[5],"framework":[6],"that":[7,74,116],"executes":[8],"multiple":[9,64],"Parameter-Server":[10],"ML":[11,91],"training":[12,92],"jobs":[13,27,65],"together":[14],"to":[15,32,36,49,82,103,112,120],"improve":[16],"cluster":[17,77],"resource":[18,30,78],"utilization.":[19],"Harmony":[20,56,75],"coordinates":[21],"fine-grained":[23],"execution":[24,69],"of":[25,53,88],"co-located":[26],"with":[28,66],"complementary":[29],"usages":[31],"avoid":[33],"contention":[34],"and":[35,98],"efficiently":[37],"share":[38],"resources":[39,119],"between":[40],"the":[41,45,50,67,89,100,113],"jobs.":[42],"To":[43],"resolve":[44],"memory":[46],"pressure":[47],"due":[48],"increased":[51],"number":[52],"simultaneous":[54],"jobs,":[55,107],"uses":[57],"data":[59],"spill/reload":[60],"mechanism":[61],"optimized":[62],"for":[63],"iterative":[68],"pattern.":[70],"Our":[71],"evaluation":[72],"shows":[73],"improves":[76],"utilization":[79],"by":[80,95,108],"up":[81],"1.65\u00d7,":[83],"resulting":[84],"in":[85],"reduction":[87],"mean":[90],"job":[93],"time":[94,102],"about":[96,109],"53%,":[97],"makespan,":[99],"total":[101],"process":[104],"all":[105],"given":[106],"38%,":[110],"compared":[111],"traditional":[114],"approaches":[115],"allocate":[117],"dedicated":[118],"each":[121],"job.":[122]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":1}],"updated_date":"2026-10-05T09:50:58.205359","created_date":"2025-10-10T00:00:00"}