{"id":"https://openalex.org/W7168037266","doi":"https://doi.org/10.1145/3806645.3816128","title":"CoDL: A Framework for Studying Cross-Component Interference in Deep Learning Training Pipelines","display_name":"CoDL: A Framework for Studying Cross-Component Interference in Deep Learning Training Pipelines","publication_year":2026,"publication_date":"2026-07-11","ids":{"openalex":"https://openalex.org/W7168037266","doi":"https://doi.org/10.1145/3806645.3816128"},"language":null,"primary_location":{"id":"doi:10.1145/3806645.3816128","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3806645.3816128","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 35th International Symposium on High-Performance Parallel and Distributed Computing","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/A5140484733","display_name":"Druva Dhakshinamoorthy","orcid":"https://orcid.org/0009-0005-5466-8473"},"institutions":[{"id":"https://openalex.org/I4210148827","display_name":"Birla Institute of Technology and Science, Pilani - Goa Campus","ror":"https://ror.org/046sh6j17","country_code":"IN","type":"education","lineage":["https://openalex.org/I4210148827","https://openalex.org/I74796645"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Druva Dhakshinamoorthy","raw_affiliation_strings":["BITS Pilani, KK Birla Goa Campus, Sancaole, India"],"raw_orcid":"https://orcid.org/0009-0005-5466-8473","affiliations":[{"raw_affiliation_string":"BITS Pilani, KK Birla Goa Campus, Sancaole, India","institution_ids":["https://openalex.org/I4210148827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090459054","display_name":"Ray A. O. Sinurat","orcid":"https://orcid.org/0000-0002-7255-7647"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ray A. O. Sinurat","raw_affiliation_strings":["The University of Chicago, Chicago, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-7255-7647","affiliations":[{"raw_affiliation_string":"The University of Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066306669","display_name":"Nikoli Dryden","orcid":null},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nikoli Dryden","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-9965-3647","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, CA, USA","institution_ids":["https://openalex.org/I1282311441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009265929","display_name":"Arnab K. Paul","orcid":"https://orcid.org/0000-0002-3694-5511"},"institutions":[{"id":"https://openalex.org/I4210148827","display_name":"Birla Institute of Technology and Science, Pilani - Goa Campus","ror":"https://ror.org/046sh6j17","country_code":"IN","type":"education","lineage":["https://openalex.org/I4210148827","https://openalex.org/I74796645"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Arnab K. Paul","raw_affiliation_strings":["BITS Pilani, KK Birla Goa Campus, Sancaole, India"],"raw_orcid":"https://orcid.org/0000-0002-3694-5511","affiliations":[{"raw_affiliation_string":"BITS Pilani, KK Birla Goa Campus, Sancaole, India","institution_ids":["https://openalex.org/I4210148827"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082476581","display_name":"Hariharan Devarajan","orcid":"https://orcid.org/0000-0001-5625-3494"},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hariharan Devarajan","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, CA, USA"],"raw_orcid":"https://orcid.org/0000-0001-5625-3494","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, CA, USA","institution_ids":["https://openalex.org/I1282311441"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"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":0.93769752,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"807","last_page":"816"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.5134999752044678,"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":0.5134999752044678,"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":0.17409999668598175,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.05649999901652336,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.7717000246047974},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.6154999732971191},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5631999969482422},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.45969998836517334},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4174000024795532},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.40610000491142273},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.4049000144004822},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.3995000123977661},{"id":"https://openalex.org/keywords/tracing","display_name":"Tracing","score":0.3774000108242035}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8083000183105469},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.7717000246047974},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.6154999732971191},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5631999969482422},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4936000108718872},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.45969998836517334},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.4207000136375427},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4174000024795532},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.40610000491142273},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.4049000144004822},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3995000123977661},{"id":"https://openalex.org/C138673069","wikidata":"https://www.wikidata.org/wiki/Q322229","display_name":"Tracing","level":2,"score":0.3774000108242035},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.3716000020503998},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3714999854564667},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.34790000319480896},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33230000734329224},{"id":"https://openalex.org/C32022120","wikidata":"https://www.wikidata.org/wiki/Q797225","display_name":"Interference (communication)","level":3,"score":0.32690000534057617},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.31940001249313354},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3149000108242035},{"id":"https://openalex.org/C2775941552","wikidata":"https://www.wikidata.org/wiki/Q25212305","display_name":"Isolation (microbiology)","level":2,"score":0.3093999922275543},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.3050000071525574},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3034000098705292},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.30230000615119934},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2815999984741211},{"id":"https://openalex.org/C2775937380","wikidata":"https://www.wikidata.org/wiki/Q1232589","display_name":"Replica","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C153740404","wikidata":"https://www.wikidata.org/wiki/Q671224","display_name":"Data center","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.25690001249313354},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.2522999942302704}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3806645.3816128","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3806645.3816128","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 35th International Symposium on High-Performance Parallel and Distributed Computing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1545880915","https://openalex.org/W2010628428","https://openalex.org/W2101778912","https://openalex.org/W2136434791","https://openalex.org/W2952046647","https://openalex.org/W2969388332","https://openalex.org/W2972087877","https://openalex.org/W2979639293","https://openalex.org/W3008805724","https://openalex.org/W3046624992","https://openalex.org/W3049494695","https://openalex.org/W3086105743","https://openalex.org/W3138303811","https://openalex.org/W3163165976","https://openalex.org/W3169404295","https://openalex.org/W3196732841","https://openalex.org/W3197816522","https://openalex.org/W3206418153","https://openalex.org/W4220938834","https://openalex.org/W4226057994","https://openalex.org/W4246569695","https://openalex.org/W4281657584","https://openalex.org/W4388031348","https://openalex.org/W4388041378","https://openalex.org/W4394998532","https://openalex.org/W4401765190","https://openalex.org/W4405755678","https://openalex.org/W4413726699","https://openalex.org/W7155557975"],"related_works":[],"abstract_inverted_index":{"Deep":[0],"learning":[1],"training":[2,15],"on":[3,42,78,138],"HPC":[4],"systems":[5],"exhibits":[6],"significant":[7],"step-time":[8],"variability\u2014data":[9],"stalls":[10],"can":[11],"dominate":[12],"50\u201385%":[13],"of":[14,102],"time":[16],"[12,":[17],"31]\u2014that":[18],"limits":[19],"efficiency.":[20],"Existing":[21],"benchmarks":[22,29],"and":[23,38,59,115],"profilers":[24,35],"isolate":[25],"individual":[26],"components:":[27],"I/O":[28],"replace":[30],"compute":[31,34,105],"with":[32,75,126],"sleep,":[33],"ignore":[36],"I/O,":[37],"GPU":[39],"tracers":[40],"operate":[41],"separate":[43],"timescales.":[44],"This":[45],"isolation":[46],"prevents":[47],"understanding":[48],"real":[49],"variability,":[50],"which":[51],"stems":[52],"from":[53,88],"interactions":[54],"among":[55],"data":[56,113,127],"loading,":[57],"compute,":[58],"communication":[60],"competing":[61],"for":[62],"shared":[63],"CPU":[64],"resources.":[65],"We":[66],"present":[67],"CoDL,":[68],"a":[69,79,97],"framework":[70],"that":[71,85],"combines":[72],"component-isolation":[73],"benchmarking":[74],"multi-level":[76],"tracing":[77],"unified":[80],"microsecond":[81],"timeline,":[82],"enabling":[83],"diagnosis":[84],"reframes":[86],"optimization":[87],"isolated":[89],"component":[90],"throughput":[91],"to":[92],"CPU\u2013GPU":[93],"scheduling":[94],"coordination.":[95],"On":[96],"single":[98],"4-GPU":[99],"node,":[100],"evaluation":[101],"ResNet-50":[103],"shows":[104],"duration":[106],"more":[107],"than":[108],"doubles":[109],"(+127.4%)":[110],"under":[111],"active":[112],"loading":[114],"NCCL":[116],"AllReduce":[117],"degrades":[118],"2.8":[119],"\u00d7,":[120],"despite":[121],"zero":[122],"direct":[123],"GPU-kernel":[124],"overlap":[125],"loading;":[128],"the":[129,139],"dominant":[130],"mechanism":[131],"is":[132],"data-loading":[133],"workers":[134],"preempting":[135],"GPU-launch":[136],"threads":[137],"CPU,":[140],"inflating":[141],"CUDA":[142],"runtime":[143],"API":[144],"latency":[145],"by":[146],"3.3":[147],"\u00d7.":[148]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-12T00:00:00"}