XENON AND NVIDIA GPUS ACCELERATE DATA SCIENCE EXECUTIVE SUMMARY Data-science workflows have traditionally been slow and cumbersome, relying on central processing units (CPUs) to load, filter, and manipulate data, and train and deploy models. Graphics processing units (GPUs) substantially reduce infrastructure costs and provide superior performance for end-to-end data-science workflows using RAPIDSTM open-source software libraries. GPU-accelerated data science can be performed on a laptop, in a data centre, and in the cloud. A single GPU server node can take on the workload of 100 CPU server nodes, which means that replacing CPU-based clusters with GPU-based clusters can accelerate data-science workloads by more than 100x while reducing operational costs and infrastructure complexi

Join for free to read