{"id":"https://openalex.org/W7224760714","doi":"https://doi.org/10.1186/s13677-026-01007-z","title":"PUCC: a reproducible cloud-based framework for optimizing deep learning input pipelines and multi-GPU utilization","display_name":"PUCC: a reproducible cloud-based framework for optimizing deep learning input pipelines and multi-GPU utilization","publication_year":2026,"publication_date":"2026-10-30","ids":{"openalex":"https://openalex.org/W7224760714","doi":"https://doi.org/10.1186/s13677-026-01007-z"},"language":"en","primary_location":{"id":"doi:10.1186/s13677-026-01007-z","doi":"https://doi.org/10.1186/s13677-026-01007-z","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13677-026-01007-z","pdf_url":null,"source":{"id":"https://openalex.org/S2486819371","display_name":"Journal of Cloud Computing Advances Systems and Applications","issn_l":"2192-113X","issn":["2192-113X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","dongbi-d","jufo-1","ki-jl-1","norway-1","oarl","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Cloud Computing","raw_type":"journal-article"},"volume":null,"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1186/s13677-026-01007-z","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5144286056","display_name":"Hamza Shahbaz","orcid":"https://orcid.org/0009-0005-7552-1585","observed_orcids":["https://orcid.org/0009-0005-7552-1585"]},"institutions":[{"id":"https://openalex.org/I194348271","display_name":"Government College University, Faisalabad","ror":"https://ror.org/051zgra59","country_code":"PK","type":"education","lineage":["https://openalex.org/I194348271"]}],"countries":["PK"],"is_corresponding":true,"raw_author_name":"Hamza Shahbaz","raw_affiliation_strings":["Department of Computer Science, Government College University Faisalabad, Faisalabad, Pakistan"],"raw_orcid":"https://orcid.org/0009-0005-7552-1585","affiliations":[{"raw_affiliation_string":"Department of Computer Science, Government College University Faisalabad, Faisalabad, Pakistan","institution_ids":["https://openalex.org/I194348271"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5133934463","display_name":"Ruhma Shahbaz","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruhma Shahbaz","raw_affiliation_strings":["Department of Vision Science, Faisalabad Medical University, Faisalabad, Pakistan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Vision Science, Faisalabad Medical University, Faisalabad, Pakistan","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5144286056"],"corresponding_institution_ids":["https://openalex.org/I194348271"],"apc_list":{"value":1790,"currency":"USD","value_usd":1790},"apc_paid":{"value":1790,"currency":"USD","value_usd":1790},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87430117,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8705999851226807,"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":0.8705999851226807,"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":0.11710000038146973,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0020000000949949026,"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/gpu-utilization","display_name":"GPU utilization","score":0.7919999957084656},{"id":"https://openalex.org/keywords/cloud-based-deep-learning","display_name":"cloud-based deep learning","score":0.7699999809265137},{"id":"https://openalex.org/keywords/multi-gpu-training","display_name":"multi-GPU training","score":0.8270000219345093},{"id":"https://openalex.org/keywords/reproducibility","display_name":"reproducibility","score":0.5669999718666077},{"id":"https://openalex.org/keywords/training-throughput","display_name":"training throughput","score":0.4180000126361847}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.75},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.7353000044822693},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.671999990940094},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.6712999939918518},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5494999885559082},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.5473999977111816},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.5153999924659729},{"id":"https://openalex.org/C41661131","wikidata":"https://www.wikidata.org/wiki/Q220764","display_name":"Interrupt","level":3,"score":0.45669999718666077},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4296000003814697},{"id":"https://openalex.org/C2779466056","wikidata":"https://www.wikidata.org/wiki/Q107630651","display_name":"Time point","level":2,"score":0.39660000801086426},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.39469999074935913},{"id":"https://openalex.org/C2776175482","wikidata":"https://www.wikidata.org/wiki/Q1195816","display_name":"Transfer (computing)","level":2,"score":0.3853999972343445},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.37860000133514404},{"id":"https://openalex.org/C2778692605","wikidata":"https://www.wikidata.org/wiki/Q4041866","display_name":"Joins","level":2,"score":0.37369999289512634},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.3659999966621399},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.36090001463890076},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.3346000015735626},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.32249999046325684},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.31929999589920044},{"id":"https://openalex.org/C19012869","wikidata":"https://www.wikidata.org/wiki/Q578372","display_name":"Response time","level":2,"score":0.3122999966144562},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C9893847","wikidata":"https://www.wikidata.org/wiki/Q1425625","display_name":"Reproducibility","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.29249998927116394},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.28450000286102295},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2808000147342682},{"id":"https://openalex.org/C81293917","wikidata":"https://www.wikidata.org/wiki/Q4189534","display_name":"System deployment","level":3,"score":0.2800999879837036},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.2782999873161316},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C2983812674","wikidata":"https://www.wikidata.org/wiki/Q134808","display_name":"Delivery system","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C81877898","wikidata":"https://www.wikidata.org/wiki/Q1965787","display_name":"Network monitoring","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.26330000162124634},{"id":"https://openalex.org/C172658912","wikidata":"https://www.wikidata.org/wiki/Q661613","display_name":"Batch processing","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C67212190","wikidata":"https://www.wikidata.org/wiki/Q104851","display_name":"Firmware","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1186/s13677-026-01007-z","doi":"https://doi.org/10.1186/s13677-026-01007-z","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13677-026-01007-z","pdf_url":null,"source":{"id":"https://openalex.org/S2486819371","display_name":"Journal of Cloud Computing Advances Systems and Applications","issn_l":"2192-113X","issn":["2192-113X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","dongbi-d","jufo-1","ki-jl-1","norway-1","oarl","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Cloud Computing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1186/s13677-026-01007-z","doi":"https://doi.org/10.1186/s13677-026-01007-z","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13677-026-01007-z","pdf_url":null,"source":{"id":"https://openalex.org/S2486819371","display_name":"Journal of Cloud Computing Advances Systems and Applications","issn_l":"2192-113X","issn":["2192-113X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","dongbi-d","jufo-1","ki-jl-1","norway-1","oarl","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Cloud Computing","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.4943999946117401,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7156910901","https://openalex.org/W4323343904","https://openalex.org/W6945344853","https://openalex.org/W7155494105","https://openalex.org/W4367360137","https://openalex.org/W2308799417","https://openalex.org/W4411351742","https://openalex.org/W4406036375","https://openalex.org/W4367147677","https://openalex.org/W4308758967"],"abstract_inverted_index":{"Abstract":[0],"Cloud-based":[1],"deep-learning":[2],"systems":[3],"can":[4],"underuse":[5],"allocated":[6,199],"GPUs":[7],"when":[8],"storage":[9],"access,":[10],"preprocessing,":[11],"and":[12,71,99,105,117,140,159,178,206,215],"host-to-device":[13],"transfer":[14],"interrupt":[15],"model":[16,37],"execution.":[17],"This":[18],"study":[19],"evaluates":[20],"the":[21,36,47,72,83,125,182],"Peak":[22,167],"Utilization":[23],"of":[24],"Cloud":[25],"Computing":[26],"(PUCC)":[27],"framework":[28],"for":[29],"improving":[30],"node-level":[31],"data":[32],"delivery":[33,57,63],"without":[34],"changing":[35],"or":[38],"adding":[39],"hardware.":[40],"PUCC":[41,190],"combines":[42],"DUVRAM":[43],"host-tensor":[44],"staging":[45],"with":[46,61,67,75],"HGPU+":[48],"execution":[49],"pipeline.":[50],"Four":[51],"controlled":[52],"conditions":[53],"were":[54,122],"tested:":[55],"generator":[56,62],"(A0),":[58],"staged":[59,65],"tensors":[60,66],"(A1),":[64],"tf.data":[68],"batching":[69],"(A2),":[70],"complete":[73],"pipeline":[74],"AUTOTUNE":[76],"prefetching":[77],"(A3).":[78],"The":[79,201],"primary":[80,126],"experiment":[81],"used":[82],"same":[84],"two":[85],"NVIDIA":[86],"Tesla":[87,115],"T4":[88],"GPUs,":[89],"DRDM1":[90],"split,":[91],"EfficientNetB0\u2013Vision":[92],"Transformer,":[93],"initialization,":[94],"optimizer,":[95],"global":[96],"batch":[97],"size,":[98],"five-epoch":[100],"workload.":[101],"Ten":[102],"matched":[103],"runs":[104],"external":[106,179],"validation":[107,208],"on":[108,198],"a":[109,118,137,187],"dual-T4":[110],"Colab":[111],"Pro":[112],"runtime,":[113],"one":[114],"T4,":[116],"four-T4":[119],"multi-node":[120],"system":[121],"completed.":[123],"In":[124],"endpoint":[127],"run,":[128],"fit":[129],"time":[130],"decreased":[131,153],"from":[132,146,154,163,172],"33.96":[133],"to":[134,148,156,165,174],"8.54":[135],"min,":[136],"74.9%":[138],"reduction":[139],"3.98-fold":[141],"speedup.":[142],"Reconstructable":[143],"throughput":[144],"increased":[145,162,171],"16.56":[147],"65.85":[149],"images/s,":[150],"step":[151],"latency":[152],"1523.3":[155],"732.0":[157],"ms,":[158],"GPU":[160,170],"utilization":[161],"45.3%":[164],"97.9%.":[166],"VRAM":[168],"per":[169],"4.15":[173],"11.05":[175],"GB.":[176],"Ten-run":[177],"analyses":[180],"reproduced":[181],"performance":[183],"direction.":[184],"Accuracy":[185],"remained":[186],"secondary":[188],"outcome.":[189],"improved":[191],"input-bound":[192],"training":[193],"by":[194],"reducing":[195],"batch-delivery":[196],"delays":[197],"GPUs.":[200],"completed":[202],"ablation,":[203],"repeated-run":[204],"analysis,":[205],"cross-platform":[207],"support":[209],"its":[210],"reproducibility":[211],"while":[212],"defining":[213],"memory":[214],"deployment":[216],"boundaries.":[217]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-10-31T00:00:00"}