{"id":"https://openalex.org/W3213026251","doi":"https://doi.org/10.1109/tnet.2021.3112082","title":"Machine Learning on Volatile Instances: Convergence, Runtime, and Cost Tradeoffs","display_name":"Machine Learning on Volatile Instances: Convergence, Runtime, and Cost Tradeoffs","publication_year":2021,"publication_date":"2021-11-10","ids":{"openalex":"https://openalex.org/W3213026251","doi":"https://doi.org/10.1109/tnet.2021.3112082","mag":"3213026251"},"language":"en","primary_location":{"id":"doi:10.1109/tnet.2021.3112082","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnet.2021.3112082","pdf_url":null,"source":{"id":"https://openalex.org/S62238642","display_name":"IEEE/ACM Transactions on Networking","issn_l":"1063-6692","issn":["1063-6692","1558-2566"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Networking","raw_type":"journal-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/A5100763625","display_name":"Xiaoxi Zhang","orcid":"https://orcid.org/0000-0003-0751-2773"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Xiaoxi Zhang","raw_affiliation_strings":["School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, Guangdong, China"],"raw_orcid":"https://orcid.org/0000-0003-0751-2773","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, Guangdong, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100398977","display_name":"Jianyu Wang","orcid":"https://orcid.org/0000-0002-7075-9333"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jianyu Wang","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0002-7075-9333","affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047067438","display_name":"Li-Feng Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li-Feng Lee","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100802745","display_name":"Tom Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tom Yang","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039389411","display_name":"Akansha Kalra","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Akansha Kalra","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067441201","display_name":"Gauri Joshi","orcid":"https://orcid.org/0000-0002-6372-9697"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gauri Joshi","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0002-6372-9697","affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085243096","display_name":"Carlee Joe\u2010Wong","orcid":"https://orcid.org/0000-0003-0785-9291"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Carlee Joe-Wong","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0003-0785-9291","affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100763625"],"corresponding_institution_ids":["https://openalex.org/I157773358"],"apc_list":null,"apc_paid":null,"fwci":0.8399,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.79371439,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":"30","issue":"1","first_page":"215","last_page":"228"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998000264167786,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998000264167786,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9990000128746033,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9979000091552734,"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":0.8974783420562744},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.747509777545929},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.7424986958503723},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.7181442379951477},{"id":"https://openalex.org/keywords/preemption","display_name":"Preemption","score":0.6042705178260803},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.5959824323654175},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5177978277206421},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5019044876098633},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46485579013824463},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.42195001244544983},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11125251650810242},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.09501096606254578}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8974783420562744},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.747509777545929},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.7424986958503723},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.7181442379951477},{"id":"https://openalex.org/C206952183","wikidata":"https://www.wikidata.org/wiki/Q1193100","display_name":"Preemption","level":2,"score":0.6042705178260803},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5959824323654175},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5177978277206421},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5019044876098633},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46485579013824463},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42195001244544983},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11125251650810242},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.09501096606254578},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tnet.2021.3112082","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnet.2021.3112082","pdf_url":null,"source":{"id":"https://openalex.org/S62238642","display_name":"IEEE/ACM Transactions on Networking","issn_l":"1063-6692","issn":["1063-6692","1558-2566"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Networking","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Decent work and economic growth","score":0.46000000834465027,"id":"https://metadata.un.org/sdg/8"}],"awards":[{"id":"https://openalex.org/G2518989827","display_name":null,"funder_award_id":"CNS-1751075","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8131504108","display_name":null,"funder_award_id":"CNS-1909306","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8529634054","display_name":null,"funder_award_id":"CCF-1850029","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/F4320307762","display_name":"International Business Machines Corporation","ror":"https://ror.org/05hh8d621"},{"id":"https://openalex.org/F4320308258","display_name":"Qualcomm","ror":"https://ror.org/002zrf773"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W1982063824","https://openalex.org/W1994616650","https://openalex.org/W2033091645","https://openalex.org/W2057950931","https://openalex.org/W2061849602","https://openalex.org/W2082819362","https://openalex.org/W2121966774","https://openalex.org/W2130062883","https://openalex.org/W2486341861","https://openalex.org/W2521140588","https://openalex.org/W2535838896","https://openalex.org/W2605643718","https://openalex.org/W2755443794","https://openalex.org/W2951436782","https://openalex.org/W2962208489","https://openalex.org/W2963248893","https://openalex.org/W2963433607","https://openalex.org/W2963889845","https://openalex.org/W2964254394","https://openalex.org/W3047261146","https://openalex.org/W3047595056","https://openalex.org/W3080812040","https://openalex.org/W3213026251","https://openalex.org/W4210307496","https://openalex.org/W6679314259","https://openalex.org/W6684191040","https://openalex.org/W6684859321","https://openalex.org/W6687483927","https://openalex.org/W6703420464","https://openalex.org/W6713134421","https://openalex.org/W6714454268","https://openalex.org/W6728757088","https://openalex.org/W6733414069","https://openalex.org/W6739622702","https://openalex.org/W6747481501","https://openalex.org/W6749257597","https://openalex.org/W6753492755","https://openalex.org/W6755251367","https://openalex.org/W6762702220"],"related_works":["https://openalex.org/W2052094875","https://openalex.org/W2249564760","https://openalex.org/W3023082165","https://openalex.org/W2128417237","https://openalex.org/W143242780","https://openalex.org/W2794408857","https://openalex.org/W2580692153","https://openalex.org/W2213644644","https://openalex.org/W1782447502","https://openalex.org/W1629725936"],"abstract_inverted_index":{"Due":[0],"to":[1,21,66,96,118,139],"the":[2,6,86,94,101,116,120,130],"massive":[3],"size":[4],"of":[5,60,88,103,110,129],"neural":[7],"network":[8],"models":[9],"and":[10,115,133,157],"training":[11,134,170],"datasets":[12],"used":[13],"in":[14,100],"machine":[15],"learning":[16],"today,":[17],"it":[18,48],"is":[19,93],"imperative":[20],"distribute":[22],"stochastic":[23],"gradient":[24,33],"descent":[25],"(SGD)":[26],"by":[27,81],"splitting":[28],"up":[29],"tasks":[30],"such":[31,54,151],"as":[32,55,152],"evaluation":[34],"across":[35],"multiple":[36],"worker":[37,105],"nodes.":[38],"However,":[39],"running":[40],"distributed":[41,145],"SGD":[42,113,146],"can":[43],"be":[44,79],"prohibitively":[45],"expensive":[46],"because":[47],"may":[49,78],"require":[50],"specialized":[51],"computing":[52],"resources":[53],"GPUs":[56],"for":[57,143],"extended":[58],"periods":[59],"time.":[61],"We":[62],"propose":[63],"cost-effective":[64],"strategies":[65,142,167],"exploit":[67],"volatile":[68,149],"cloud":[69,160],"instances":[70,150,156],"that":[71,165],"are":[72,137],"cheaper":[73],"than":[74],"standard":[75],"instances,":[76,131],"but":[77],"interrupted":[80],"higher":[82],"priority":[83],"workloads.":[84],"To":[85],"best":[87],"our":[89,166],"knowledge,":[90],"this":[91],"work":[92],"first":[95],"quantify":[97],"how":[98],"variations":[99],"number":[102],"active":[104],"nodes":[106],"(as":[107],"a":[108],"result":[109],"preemption)":[111],"affect":[112],"convergence":[114],"time":[117],"train":[119],"model.":[121],"By":[122],"understanding":[123],"these":[124],"trade-offs":[125],"between":[126],"preemption":[127],"probability":[128],"accuracy,":[132],"time,":[135],"we":[136],"able":[138],"derive":[140],"practical":[141],"configuring":[144],"jobs":[147],"on":[148],"Amazon":[153],"EC2":[154],"spot":[155],"other":[158],"preemptible":[159],"instances.":[161],"Experimental":[162],"results":[163],"show":[164],"achieve":[168],"good":[169],"performance":[171],"at":[172],"substantially":[173],"lower":[174],"cost.":[175]},"counts_by_year":[{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}