{"id":"W4415657680","doi":"10.1038/s41597-026-07496-6","title":"Characterization of high-resolution AI data center training workloads on single and multiple GPU nodes","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Cloud Computing and Resource Management","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University; Independent Electricity System Operator","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Session (web analytics); Resilience (materials science); Grid; Data center; Set (abstract data type); Energy consumption; Data set; Training set; Training (meteorology)","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003986077,0.0006027559,0.0003994984,0.0007844346,0.0004591761,0.0004595676,0.0007201047,0.0004643894,0.000930032],"category_scores_gemma":[0.001776294,0.0001713409,0.0002825263,0.0009300485,0.0003154501,0.0005128989,0.0005801262,0.0008192268,0.0006050605],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004623664,"about_ca_system_score_gemma":0.0005094174,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007943722,"about_ca_topic_score_gemma":0.01688323,"domain_scores_codex":[0.9995401,0.00005046684,0.00002793712,0.0001474795,0.0001199585,0.0001141519],"domain_scores_gemma":[0.9992535,0.0001467434,0.00005485522,0.0001680362,0.0002520879,0.0001247984],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.003018132,0.003926326,0.3486632,0.0008612947,0.0004346438,0.001286407,0.001331555,0.2086498,0.08714094,0.003076163,0.1502704,0.1913412],"study_design_scores_gemma":[0.0001400185,0.00070297,0.4311088,0.00006969718,0.00007059721,0.0006856957,0.0009127672,0.512261,0.02728926,0.001993323,0.02467725,0.00008863407],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9652938,0.0003276421,0.008785044,0.0003529561,0.0001860926,0.0001404687,0.01872124,0.002056939,0.00413594],"genre_scores_gemma":[0.9480125,0.0001412549,0.006243196,0.000121779,0.00006496929,0.0001032345,0.04392627,0.0001213577,0.001265368],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007943722,"threshold_uncertainty_score":0.01579499,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06182922118262564,"score_gpt":0.2691192533714195,"score_spread":0.2072900321887939,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}