{"id":"https://openalex.org/W4297882618","doi":"https://doi.org/10.1145/3492866.3549716","title":"On scheduling ring-all-reduce learning jobs in multi-tenant GPU clusters with communication contention","display_name":"On scheduling ring-all-reduce learning jobs in multi-tenant GPU clusters with communication contention","publication_year":2022,"publication_date":"2022-09-21","ids":{"openalex":"https://openalex.org/W4297882618","doi":"https://doi.org/10.1145/3492866.3549716"},"language":"en","primary_location":{"id":"doi:10.1145/3492866.3549716","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3492866.3549716","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://vtechworks.lib.vt.edu/bitstreams/2b26fdb1-60bd-41ad-9e54-37b63cf69502/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031558801","display_name":"Menglu Yu","orcid":"https://orcid.org/0000-0002-3653-3453"},"institutions":[{"id":"https://openalex.org/I173911158","display_name":"Iowa State University","ror":"https://ror.org/04rswrd78","country_code":"US","type":"education","lineage":["https://openalex.org/I173911158"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Menglu Yu","raw_affiliation_strings":["Iowa State University"],"affiliations":[{"raw_affiliation_string":"Iowa State University","institution_ids":["https://openalex.org/I173911158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040279482","display_name":"Bo Ji","orcid":"https://orcid.org/0000-0003-0149-7509"},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bo Ji","raw_affiliation_strings":["Virginia Tech"],"affiliations":[{"raw_affiliation_string":"Virginia Tech","institution_ids":["https://openalex.org/I859038795"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059626072","display_name":"Hridesh Rajan","orcid":"https://orcid.org/0000-0002-9410-9562"},"institutions":[{"id":"https://openalex.org/I173911158","display_name":"Iowa State University","ror":"https://ror.org/04rswrd78","country_code":"US","type":"education","lineage":["https://openalex.org/I173911158"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hridesh Rajan","raw_affiliation_strings":["Iowa State University"],"affiliations":[{"raw_affiliation_string":"Iowa State University","institution_ids":["https://openalex.org/I173911158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100409661","display_name":"Jia Liu","orcid":"https://orcid.org/0000-0001-8844-3233"},"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"]},{"id":"https://openalex.org/I173911158","display_name":"Iowa State University","ror":"https://ror.org/04rswrd78","country_code":"US","type":"education","lineage":["https://openalex.org/I173911158"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jia Liu","raw_affiliation_strings":["The Ohio State University and Iowa State University"],"affiliations":[{"raw_affiliation_string":"The Ohio State University and Iowa State University","institution_ids":["https://openalex.org/I52357470","https://openalex.org/I173911158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5031558801"],"corresponding_institution_ids":["https://openalex.org/I173911158"],"apc_list":null,"apc_paid":null,"fwci":1.427,"has_fulltext":true,"cited_by_count":10,"citation_normalized_percentile":{"value":0.82622258,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"21","last_page":"30"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9991000294685364,"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"}},"topics":[{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9991000294685364,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9987999796867371,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9986000061035156,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.8441794514656067},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8219811916351318},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6307762265205383},{"id":"https://openalex.org/keywords/job-shop-scheduling","display_name":"Job shop scheduling","score":0.6260491609573364},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.553022563457489},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5387127995491028},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5095677375793457},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47783374786376953},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.4630333185195923},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.40681135654449463},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35909503698349},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3412139117717743},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.20631346106529236},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.2004346251487732},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.15111318230628967},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.1468256711959839},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.10471728444099426}],"concepts":[{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.8441794514656067},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8219811916351318},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6307762265205383},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.6260491609573364},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.553022563457489},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5387127995491028},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5095677375793457},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47783374786376953},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.4630333185195923},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.40681135654449463},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35909503698349},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3412139117717743},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.20631346106529236},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2004346251487732},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.15111318230628967},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.1468256711959839},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.10471728444099426},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"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}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3492866.3549716","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3492866.3549716","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing","raw_type":"proceedings-article"},{"id":"pmh:oai:vtechworks.lib.vt.edu:10919/112236","is_oa":true,"landing_page_url":"http://hdl.handle.net/10919/112236","pdf_url":"https://vtechworks.lib.vt.edu/bitstreams/2b26fdb1-60bd-41ad-9e54-37b63cf69502/download","source":{"id":"https://openalex.org/S4306400248","display_name":"VTechWorks (Virginia Tech)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I859038795","host_organization_name":"Virginia Tech","host_organization_lineage":["https://openalex.org/I859038795"],"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":"Text"}],"best_oa_location":{"id":"pmh:oai:vtechworks.lib.vt.edu:10919/112236","is_oa":true,"landing_page_url":"http://hdl.handle.net/10919/112236","pdf_url":"https://vtechworks.lib.vt.edu/bitstreams/2b26fdb1-60bd-41ad-9e54-37b63cf69502/download","source":{"id":"https://openalex.org/S4306400248","display_name":"VTechWorks (Virginia Tech)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I859038795","host_organization_name":"Virginia Tech","host_organization_lineage":["https://openalex.org/I859038795"],"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":"Text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1556298747","display_name":null,"funder_award_id":"CAREER CNS-2110259","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G2283100477","display_name":null,"funder_award_id":"CNS-2110259","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3071850680","display_name":null,"funder_award_id":"2110259","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4572141212","display_name":null,"funder_award_id":"CCF-2110252","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G486605795","display_name":null,"funder_award_id":"211247","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6139389680","display_name":"NeTS: Small: Toward Optimal, Efficient, and Holistic Networking Design for Massive-MIMO Wireless Networks","funder_award_id":"2102233","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6614458066","display_name":"CIF: Small: Taming Convergence and Delay in Stochastic Network Optimization with Hessian Information","funder_award_id":"2110252","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6671297155","display_name":null,"funder_award_id":"CAREER","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G848032724","display_name":null,"funder_award_id":"Science","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8852016375","display_name":null,"funder_award_id":"CNS-2102233","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320307791","display_name":"Cisco Systems","ror":"https://ror.org/03yt1ez60"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4297882618.pdf","grobid_xml":"https://content.openalex.org/works/W4297882618.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W2034008996","https://openalex.org/W2057332538","https://openalex.org/W2612387305","https://openalex.org/W2614112306","https://openalex.org/W2734941459","https://openalex.org/W2760837370","https://openalex.org/W2798515322","https://openalex.org/W2919594608","https://openalex.org/W2964324519","https://openalex.org/W2982346115","https://openalex.org/W3187104792","https://openalex.org/W3195168961","https://openalex.org/W4221149636","https://openalex.org/W4241618768","https://openalex.org/W6600001191","https://openalex.org/W6600281463","https://openalex.org/W6607094058","https://openalex.org/W6713134421"],"related_works":["https://openalex.org/W2595172197","https://openalex.org/W2084856301","https://openalex.org/W2127970246","https://openalex.org/W2885125400","https://openalex.org/W1989889224","https://openalex.org/W4382618745","https://openalex.org/W2895982276","https://openalex.org/W2039885954","https://openalex.org/W2918887334","https://openalex.org/W2973058717"],"abstract_inverted_index":{"Powered":[0],"by":[1],"advances":[2],"in":[3,73,151,220],"deep":[4],"learning":[5,9],"(DL)":[6],"techniques,":[7],"machine":[8],"and":[10,28,70,179,254,256,283],"artificial":[11],"intelligence":[12],"have":[13,57],"achieved":[14],"astonishing":[15],"successes.":[16],"However,":[17,98],"the":[18,46,52,86,94,173,177,184,192,198,207,217,233,264,276],"rapidly":[19],"growing":[20],"needs":[21],"for":[22,33,49,88,139,227],"DL":[23,35,50,102,119,141,212],"also":[24],"led":[25],"to":[26,65,116,133,147,172,183,205,262,275],"communication-":[27],"resource-intensive":[29],"distributed":[30,279],"training":[31,103,120,142,213],"jobs":[32,104],"large-scale":[34],"training,":[36,51],"which":[37,144],"are":[38,105,157],"typically":[39],"deployed":[40,106],"over":[41,107,268],"GPU":[42,74,108,280],"clusters.":[43,75],"To":[44,215],"sustain":[45],"ever-increasing":[47],"demand":[48],"so-called":[53],"\"ring-all-reduce\"":[54],"(RAR)":[55],"technologies":[56],"recently":[58],"emerged":[59],"as":[60,200],"a":[61,126,162,201],"favorable":[62],"computing":[63],"architecture":[64],"efficiently":[66],"process":[67],"network":[68],"communication":[69,96,110,169,180],"computation":[71],"load":[72],"The":[76],"most":[77],"salient":[78],"feature":[79],"of":[80,128,176,186,209,266,278],"RAR":[81],"is":[82,224],"that":[83,166,223,240],"it":[84],"removes":[85],"need":[87],"dedicated":[89],"parameter":[90],"servers,":[91],"thus":[92],"alleviating":[93],"potential":[95],"bottleneck.":[97],"when":[99],"multiple":[100],"RAR-based":[101,140,211],"clusters,":[109],"bottlenecks":[111],"could":[112],"still":[113],"occur":[114],"due":[115],"contentions":[117],"between":[118],"jobs.":[121,214],"So":[122],"far,":[123],"there":[124],"remains":[125],"lack":[127],"theoretical":[129],"understanding":[130],"on":[131,191],"how":[132],"design":[134,245],"contention-aware":[135],"resource":[136],"scheduling":[137],"algorithms":[138],"jobs,":[143],"motivates":[145],"us":[146],"fill":[148],"this":[149,152,221],"gap":[150],"work.":[153],"Our":[154],"main":[155],"contributions":[156],"three-fold:":[158],"i)":[159],"We":[160,258],"develop":[161],"new":[163],"analytical":[164,194],"model":[165],"characterizes":[167],"both":[168],"overhead":[170],"related":[171,182],"worker":[174],"distribution":[175],"job":[178],"contention":[181],"co-location":[185],"different":[187],"jobs;":[188],"ii)":[189],"Based":[190],"proposed":[193],"model,":[195],"we":[196,231],"formulate":[197],"problem":[199,222,234],"non-convex":[202],"integer":[203,237],"program":[204,239],"minimize":[206],"makespan":[208],"all":[210],"address":[216],"unique":[218],"structure":[219],"not":[225],"amenable":[226],"optimization":[228,282],"algorithm":[229,244,284],"design,":[230],"reformulate":[232],"into":[235],"an":[236],"linear":[238],"enables":[241],"provable":[242],"approximation":[243],"called":[246],"SJF-BCO":[247,267],"(Smallest":[248],"Job":[249],"First":[250],"with":[251],"Balanced":[252],"Contention":[253],"Overhead);":[255],"iii)":[257],"conduct":[259],"extensive":[260],"experiments":[261],"show":[263],"superiority":[265],"existing":[269],"schedulers.":[270],"Collectively,":[271],"our":[272],"results":[273],"contribute":[274],"state-of-the-art":[277],"system":[281],"design.":[285]},"counts_by_year":[{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1}],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2025-10-10T00:00:00"}