1、Sheng Luo !GPU#$% Data prep includes joins, variable transformations CPU Cluster Configuration CPU nodes (61 GiB memory, 8 vCPUs, 64- bit platform), Apache Spark DGX Cluster Configuration 5x DGX-1 on InfiniBand network 8762 6148 3925 3221 322 213 End-to-End 10 FASTER SPEEDS, REAL-WORLD BENEFITS cuIO
2、/cuDF Load and Data PreparationXGBoost Machine Learning Benchmark 200GB CSV dataset; Data prep includes joins, variable transformations CPU Cluster Configuration CPU nodes (61 GiB memory, 8 vCPUs, 64- bit platform), Apache Spark DGX Cluster Configuration 5x DGX-1 on InfiniBand network End-to-End Imp
3、roving Over Time Time in seconds (shorter is better) cuIO/cuDF (Load and Data Prep)Data ConversionXGBoost 11 FASTER SPEEDS, REAL-WORLD BENEFITS cuIO/cuDF Load and Data PreparationXGBoost Machine Learning Time in seconds (shorter is better) cuIO/cuDF (Load and Data Prep)Data ConversionXGBoost Benchmark 200GB CSV dataset; Data prep includes joins, variable transformations CPU Cluster Configuration C
sgpjbg002
工作日 8:30 - 17:30
报告分类
