{"id":"https://openalex.org/W7161108473","doi":"https://doi.org/10.1145/3746467.3801513","title":"Identifying the Optimal Number of DataLoader Workers for CPU-GPU Concurrency in Asynchronous Deep Training Pipelines","display_name":"Identifying the Optimal Number of DataLoader Workers for CPU-GPU Concurrency in Asynchronous Deep Training Pipelines","publication_year":2026,"publication_date":"2026-04-23","ids":{"openalex":"https://openalex.org/W7161108473","doi":"https://doi.org/10.1145/3746467.3801513"},"language":null,"primary_location":{"id":"doi:10.1145/3746467.3801513","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746467.3801513","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 ACM Southeast Conference","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746467.3801513","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129705383","display_name":"Michael Seavers","orcid":null},"institutions":[{"id":"https://openalex.org/I94339441","display_name":"Western Kentucky University","ror":"https://ror.org/0446vnd56","country_code":"US","type":"education","lineage":["https://openalex.org/I94339441"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Michael Seavers","raw_affiliation_strings":["Western Kentucky University, Bowling Green, KY, USA"],"raw_orcid":"https://orcid.org/0009-0001-7019-1979","affiliations":[{"raw_affiliation_string":"Western Kentucky University, Bowling Green, KY, USA","institution_ids":["https://openalex.org/I94339441"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5136110965","display_name":"Qi Li","orcid":"https://orcid.org/0000-0001-5554-5647"},"institutions":[{"id":"https://openalex.org/I94339441","display_name":"Western Kentucky University","ror":"https://ror.org/0446vnd56","country_code":"US","type":"education","lineage":["https://openalex.org/I94339441"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qi Li","raw_affiliation_strings":["SEAS, Western Kentucky University, Bowling Green, KY, USA"],"raw_orcid":"https://orcid.org/0000-0001-5554-5647","affiliations":[{"raw_affiliation_string":"SEAS, Western Kentucky University, Bowling Green, KY, USA","institution_ids":["https://openalex.org/I94339441"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5129705383"],"corresponding_institution_ids":["https://openalex.org/I94339441"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.96464226,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"270","last_page":"275"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.3856000006198883,"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":0.3856000006198883,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.3224000036716461,"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/T14347","display_name":"Big Data and Digital Economy","score":0.0608999989926815,"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/concurrency","display_name":"Concurrency","score":0.7235999703407288},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.6534000039100647},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6209999918937683},{"id":"https://openalex.org/keywords/provisioning","display_name":"Provisioning","score":0.5794000029563904},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5781999826431274},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5175999999046326},{"id":"https://openalex.org/keywords/pipeline-transport","display_name":"Pipeline transport","score":0.47609999775886536},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.4643000066280365},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.43779999017715454}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.847599983215332},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.7235999703407288},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.6534000039100647},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6399999856948853},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6209999918937683},{"id":"https://openalex.org/C172191483","wikidata":"https://www.wikidata.org/wiki/Q1071806","display_name":"Provisioning","level":2,"score":0.5794000029563904},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5781999826431274},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5175999999046326},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.47609999775886536},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.4643000066280365},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.43779999017715454},{"id":"https://openalex.org/C150495011","wikidata":"https://www.wikidata.org/wiki/Q128392","display_name":"Concurrent computing","level":2,"score":0.3865000009536743},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.33709999918937683},{"id":"https://openalex.org/C84511453","wikidata":"https://www.wikidata.org/wiki/Q2914952","display_name":"Concurrency control","level":3,"score":0.334199994802475},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.32910001277923584},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3215999901294708},{"id":"https://openalex.org/C149091818","wikidata":"https://www.wikidata.org/wiki/Q2429814","display_name":"Software system","level":3,"score":0.3149000108242035},{"id":"https://openalex.org/C2775941552","wikidata":"https://www.wikidata.org/wiki/Q25212305","display_name":"Isolation (microbiology)","level":2,"score":0.2806999981403351},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.28040000796318054},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.2797999978065491},{"id":"https://openalex.org/C174683762","wikidata":"https://www.wikidata.org/wiki/Q609588","display_name":"Component-based software engineering","level":4,"score":0.2768999934196472},{"id":"https://openalex.org/C2778814252","wikidata":"https://www.wikidata.org/wiki/Q5156715","display_name":"Composability","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2743000090122223},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.271699994802475},{"id":"https://openalex.org/C154556556","wikidata":"https://www.wikidata.org/wiki/Q192969","display_name":"Computer multitasking","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C35869016","wikidata":"https://www.wikidata.org/wiki/Q846636","display_name":"Software architecture","level":3,"score":0.2615000009536743},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C67953723","wikidata":"https://www.wikidata.org/wiki/Q192525","display_name":"Workstation","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746467.3801513","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746467.3801513","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 ACM Southeast Conference","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3746467.3801513","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746467.3801513","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 ACM Southeast Conference","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","score":0.6177725791931152,"display_name":"Decent work and economic growth"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W1864199185","https://openalex.org/W2886675192","https://openalex.org/W4372260356","https://openalex.org/W4394745249","https://openalex.org/W4414427843"],"related_works":[],"abstract_inverted_index":{"Efficient":[0],"CPU-GPU":[1,106,118],"concurrency":[2,140],"is":[3,21],"critical":[4],"for":[5,104],"achieving":[6],"high":[7],"performance":[8],"in":[9,144],"deep":[10,146],"learning":[11],"training":[12,147],"pipelines,":[13],"yet":[14],"selecting":[15],"the":[16,42,99,110,133],"number":[17],"of":[18,34,59],"DataLoader":[19,39,131],"workers":[20,68],"still":[22],"largely":[23],"driven":[24],"by":[25,37],"empirical":[26],"heuristics.":[27],"This":[28],"paper":[29],"presents":[30],"a":[31],"system-level":[32],"view":[33],"worker":[35],"provisioning":[36],"treating":[38],"processes":[40],"as":[41],"runtime":[43],"mechanism":[44],"that":[45],"mediates":[46],"host-side":[47],"supply":[48],"and":[49,67,81,92,124],"device-side":[50],"demand.":[51],"We":[52,108],"introduce":[53],"an":[54],"architectural":[55],"abstraction,":[56],"SM:CC:W\u2014the":[57],"ratio":[58],"GPU":[60],"Streaming":[61],"Multiprocessors":[62],"(SMs),":[63],"CPU":[64],"cores":[65],"(CC),":[66],"(W)\u2014and":[69],"interpret":[70],"how":[71,138],"it":[72],"reflects":[73],"fundamental":[74],"constraints":[75],"arising":[76],"from":[77],"concurrency,":[78],"latency":[79],"hiding,":[80],"scheduling":[82],"overhead.":[83],"By":[84],"jointly":[85],"analyzing":[86],"elapsed":[87],"time,":[88],"GPU/CPU":[89],"utilization":[90],"rates,":[91],"workload":[93],"distribution":[94],"across":[95],"major":[96],"pipeline":[97],"components,":[98],"framework":[100],"provides":[101,135],"structured":[102],"guidance":[103],"understanding":[105],"overlap.":[107],"evaluate":[109],"approach":[111],"on":[112],"five":[113],"single-machine":[114],"platforms":[115],"with":[116],"diverse":[117],"configurations,":[119],"including":[120],"three":[121],"local":[122],"systems":[123],"two":[125],"Google":[126],"Colab":[127],"environments.":[128],"Beyond":[129],"supporting":[130],"tuning,":[132],"study":[134],"insight":[136],"into":[137],"software":[139],"shapes":[141],"heterogeneous":[142],"execution":[143],"modern":[145],"systems.":[148]},"counts_by_year":[],"updated_date":"2026-05-15T06:12:33.780692","created_date":"2026-05-15T00:00:00"}