The power of RAPIDS, now available for Windows

Install Now

Use RAPIDS
on Windows

Windows users can now tap into GPU accelerated data science on their local machines using RAPIDS on Windows Subsystem for Linux2 (WSL2)! WSL2 is a Windows feature that enables users to run native Linux command-line tools directly on Windows. Using this feature does not require a dual boot environment, removing complexity and saving you time. Your computer will need an NVIDIA GPU with Compute capability 7.0 or higher.

Before Installing

Prerequisites

OS: Windows 11.

WSL Version: WSL2. WSL1 is not supported.

GPU: Only GPUs with Compute capability 7.0 or higher are supported on RAPIDS in WSL2. GPUs with 16GB or more of RAM is recommended

WSL2 Instance: Ubuntu 20.04 instance for WSL2.

Limitations

Only single GPU is supported.

GPU Direct Storage is not supported.

At least 8 GB of RAM and a relatively fast CPU are strongly recommended https://docs.microsoft.com/en-us/windows/wsl/install

Choose a Docker Installation Method

Install Docker Desktop in Windows and enable the WSL2 backend. Docker packages run in WSL2 will only work if Docker Desktop is correctly set up and running in Windows. Learn more.

Install Docker + nvidia-docker directly in WSL2 following the current instructions on the RAPIDS site. Note that WSL2 does not have systemd so Docker will not start automatically. You have to manually run sudo service docker start after each reboot. [Learn more].

Troubleshooting

When installing with conda, if an http 000 connection error occurs when accessing the repository data, run wsl --shutdown and then restart the WSL instance. More information

Connect

Join our community conversations about RAPIDS on WSL2 using Twitter, Slack, or ask a question on StackOverflow.

Installation Instructions

Conda
(Preferred Method)

  1. Install WSL2 and the Ubuntu 20.04 package using Microsoft’s instructions.
  2. Install the latest NVIDIA Drivers.
  3. Log into your WSL2 Linux instance.
  4. Install Conda on your WSL2 Linux Instance using our Conda instructions.
  5. Install RAPIDS via Conda, using the RAPIDS Release Selector tool.
  6. Install additional GitHub repositories and enablements from the Learn More section.
  7. Run this code to check that your RAPIDS installation is working:
     import cudf
     print(cudf.Series([1, 2, 3]))
    

Docker Desktop
(Docker Method #1)

  1. Install WSL2 and the Ubuntu 20.04 package using Microsoft’s instructions.
  2. Install the latest NVIDIA Drivers.
  3. Install latest Docker Desktop for Windows according to your applicable licensing terms.
  4. Log into your WSL2 Linux instance.
  5. Generate and run the RAPIDS docker pull and docker run commands based on your desired configuration using the RAPIDS Release Selector.
  6. Inside your docker instance, run this code to check that your RAPIDS installation is working:
     import cudf
     print(cudf.Series([1, 2, 3]))
    
  7. Install additional Github repos, enablements, and 3rd party tools from the Learn More section.

Docker on WSL2
(Docker Method #2)

  1. Install WSL2 and the Ubuntu 20.04 package using Microsoft’s instructions.
  2. Install the latest NVIDIA Drivers.
  3. Log into your WSL2 Linux instance.
  4. Install the Docker Environment on your WSL2 Linux instance using the Docker instructions. You may need to add sudo before any Docker command.
  5. Generate and run the RAPIDS docker pull and docker run commands based on your desired configuration using the RAPIDS Release Selector. You may need to add sudo before any Docker command.
  6. Inside your docker instance, run this code to check that your RAPIDS installation is working:
     import cudf
     print(cudf.Series([1, 2, 3]))
    
  7. Install additional Github repos, enablements, and 3rd party tools from the Learn More section.