{"id":"https://openalex.org/W7155629922","doi":"https://doi.org/10.1145/3807454","title":"Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments","display_name":"Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments","publication_year":2026,"publication_date":"2026-04-25","ids":{"openalex":"https://openalex.org/W7155629922","doi":"https://doi.org/10.1145/3807454"},"language":"en","primary_location":{"id":"doi:10.1145/3807454","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3807454","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1145/3807454","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5000069771","display_name":"Sichao Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Sichao Chen","raw_affiliation_strings":["Harbin Institute of Technology"],"raw_orcid":"https://orcid.org/0009-0002-5696-4944","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100733744","display_name":"Desheng Wang","orcid":"https://orcid.org/0000-0002-7502-7094"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Desheng Wang","raw_affiliation_strings":["Harbin Institute of Technology, Shenzhen"],"raw_orcid":"https://orcid.org/0000-0002-7502-7094","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology, Shenzhen","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085221201","display_name":"Weizhe Zhang","orcid":"https://orcid.org/0000-0003-4783-876X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weizhe Zhang","raw_affiliation_strings":["Harbin Institute of Technology"],"raw_orcid":"https://orcid.org/0000-0003-4783-876X","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060954730","display_name":"Meng Hao","orcid":"https://orcid.org/0000-0003-0043-4370"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meng Hao","raw_affiliation_strings":["Harbin Institute of Technology"],"raw_orcid":"https://orcid.org/0000-0003-0043-4370","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060251784","display_name":"Yu\u2010Chu Tian","orcid":"https://orcid.org/0000-0002-8709-5625"},"institutions":[{"id":"https://openalex.org/I160993911","display_name":"Queensland University of Technology","ror":"https://ror.org/03pnv4752","country_code":"AU","type":"education","lineage":["https://openalex.org/I160993911"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Yu-Chu Tian","raw_affiliation_strings":["Queensland University of Technology"],"raw_orcid":"https://orcid.org/0000-0002-8709-5625","affiliations":[{"raw_affiliation_string":"Queensland University of Technology","institution_ids":["https://openalex.org/I160993911"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5134589653","display_name":"Hui He","orcid":"https://orcid.org/0000-0002-6494-775X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hui He","raw_affiliation_strings":["Harbin Institute of Technology"],"raw_orcid":"https://orcid.org/0000-0002-6494-775X","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.53546443,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"23","issue":"2","first_page":"1","last_page":"25"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.28949999809265137,"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":0.28949999809265137,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.2605000138282776,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.09950000047683716,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.5766000151634216},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.5526999831199646},{"id":"https://openalex.org/keywords/performance-prediction","display_name":"Performance prediction","score":0.526199996471405},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5019000172615051},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5011000037193298},{"id":"https://openalex.org/keywords/performance-improvement","display_name":"Performance improvement","score":0.43160000443458557},{"id":"https://openalex.org/keywords/concurrency","display_name":"Concurrency","score":0.4187999963760376},{"id":"https://openalex.org/keywords/shared-resource","display_name":"Shared resource","score":0.414000004529953}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9185000061988831},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.5766000151634216},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.5526999831199646},{"id":"https://openalex.org/C2777115002","wikidata":"https://www.wikidata.org/wiki/Q7168246","display_name":"Performance prediction","level":2,"score":0.526199996471405},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5019000172615051},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5011000037193298},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.43160000443458557},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.42100000381469727},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.4187999963760376},{"id":"https://openalex.org/C51332947","wikidata":"https://www.wikidata.org/wiki/Q1172305","display_name":"Shared resource","level":2,"score":0.414000004529953},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.40529999136924744},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.40470001101493835},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3962000012397766},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.38659998774528503},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.3529999852180481},{"id":"https://openalex.org/C38725249","wikidata":"https://www.wikidata.org/wiki/Q913876","display_name":"Time-sharing","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3474000096321106},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.34049999713897705},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3160000145435333},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3100999891757965},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.2924000024795532},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.2840000092983246},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2782000005245209},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.2565000057220459}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3807454","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3807454","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3807454","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3807454","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2606722458","https://openalex.org/W2794670651","https://openalex.org/W2794729807","https://openalex.org/W2935614364","https://openalex.org/W2982157693","https://openalex.org/W2984189761","https://openalex.org/W3189983613","https://openalex.org/W4220877690","https://openalex.org/W4304192541","https://openalex.org/W4401408761","https://openalex.org/W4408867174","https://openalex.org/W4408867487","https://openalex.org/W4416203719"],"related_works":[],"abstract_inverted_index":{"GPU":[0,6,46,65,95,149,202],"sharing":[1,13,24,67,75,151],"is":[2,156],"commonly":[3],"employed":[4],"in":[5,45,64,243],"clusters":[7],"to":[8,27,35,99,182,235],"improve":[9],"utilization,":[10],"with":[11,145],"spatial":[12,23,66,74,150],"being":[14],"one":[15],"of":[16,88,135,158,163,170,240],"the":[17,113,132,160,237],"most":[18],"widely":[19],"adopted":[20],"techniques.":[21],"However,":[22],"can":[25],"lead":[26],"resource":[28,96],"interference,":[29],"making":[30],"task":[31,42,50,109,140,165],"execution":[32,38,87,133,161],"times":[33,39],"difficult":[34],"predict.":[36],"Predictable":[37],"for":[40,60,76,115,166],"each":[41,164],"are":[43],"crucial":[44],"cluster":[47],"management":[48],"and":[49,94,117,197,219,231],"scheduling.":[51],"In":[52],"this":[53],"article,":[54],"we":[55,111,125],"propose":[56],"a":[57,81,127,136,224,232],"performance":[58,93,128,225],"predictor":[59,129,155,242],"multi-DNN":[61],"training":[62,139,172],"tasks":[63,90,147],"environments.":[68,246],"We":[69,221],"first":[70],"conduct":[71],"experiments":[72],"on":[73,80,122,176,200],"multiple":[77,89,201],"DNN":[78,138,171],"workloads":[79],"single":[82],"GPU,":[83],"demonstrating":[84],"that":[85,130,180,213],"concurrent":[86],"improves":[91],"overall":[92],"utilization":[97],"compared":[98,181],"serial":[100],"execution.":[101],"By":[102],"analyzing":[103],"warp":[104],"stall":[105],"reasons":[106],"collected":[107],"during":[108],"execution,":[110],"investigate":[112],"interference":[114],"computation":[116],"memory":[118],"resources":[119],"under":[120,148,227],"MPS":[121],"GPUs.":[123],"Finally,":[124],"design":[126],"predicts":[131],"time":[134,162],"target":[137],"when":[141],"it":[142],"runs":[143],"concurrently":[144],"other":[146,183],"via":[152],"MPS.":[153],"The":[154],"capable":[157],"predicting":[159],"previously":[167],"unseen":[168],"combinations":[169],"tasks.":[173],"Extensive":[174],"evaluations":[175],"modern":[177],"GPUs":[178],"show":[179],"baseline":[184],"methods,":[185],"our":[186,214,241],"approach":[187],"exhibits":[188],"higher":[189,208],"prediction":[190],"accuracy,":[191],"as":[192,194,204,206],"well":[193,205],"improved":[195],"stability":[196],"robustness.":[198],"Experiments":[199],"architectures,":[203],"at":[207],"concurrency":[209],"levels,":[210],"further":[211],"demonstrate":[212],"method":[215],"possesses":[216],"strong":[217],"generalization":[218],"scalability.":[220],"also":[222],"conducted":[223],"analysis":[226],"diverse":[228],"workload":[229],"pattern":[230],"case":[233],"study":[234],"validate":[236],"practical":[238],"applicability":[239],"real":[244],"scheduling":[245]},"counts_by_year":[],"updated_date":"2026-06-26T06:17:10.115597","created_date":"2026-04-26T00:00:00"}