Google Cloud Platform#
Compute Engine Instance
Launch a Compute Engine instance and run CUDA-X.
single-node
Vertex AI
Launch the RAPIDS container in Vertex AI managed notebooks.
single-node
Google Kubernetes Engine (GKE)
Launch a CUDA-X cluster on managed Kubernetes.
multi-node
Dataproc
Launch a CUDA-X cluster on Dataproc.
multi-node
NVIDIA CUDA-X libraries for data science can be deployed on Google Cloud Platform in several ways. Google Cloud supports various kinds of GPU VMs for different needs. Please visit the Google Cloud documentation for an overview of GPU VM sizes and GPU VM availability by region.
Examples#
Measuring Performance with the One Billion Row Challenge
Generate one billion rows with cuDF and CuPy, then aggregate them with Dask and cuDF about 60x faster than pandas.
Time Series Forecasting on GKE
Forecast Walmart M5 sales by engineering features with cuDF and tuning XGBoost with Optuna on a GKE Dask cluster.
HPO with dask-ml and cuml
Tune XGBoost and a cuML random forest on airline data with dask-ml grid and random search across local GPUs.
Autoscaling Multi-Tenant Kubernetes Deep-Dive
Simulate many users running Dask GPU workloads on an autoscaling GKE cluster and measure wait times and GPU utilization.